AU2019220327A1

AU2019220327A1 - Control method for a hydraulic unit

Info

Publication number: AU2019220327A1
Application number: AU2019220327A
Authority: AU
Inventors: Abdelkrim Benchaib; Gilney DAMM; Bruno Luscan; Janailson RODRIGUES
Original assignee: Universite D'Evry Val D'Essonne; SuperGrid Institute SAS
Current assignee: Universite D'Evry Val D'Essonne; SuperGrid Institute SAS
Priority date: 2018-02-14
Filing date: 2019-02-13
Publication date: 2020-09-03
Anticipated expiration: 2039-02-13
Also published as: AU2019220327B2; EP3746655B1; FR3077849A1; FR3077849B1; WO2019158600A1; PT3746655T; EP3746655A1

Abstract

The invention relates to a control method for a hydraulic unit (1) comprising a variable-speed turbine (300) that drives an electric generator (210), comprising: -having a minimum characteristic rotational speed law of the turbine; -making the hydraulic unit operate with an initial electrical power setpoint value; -retrieving a value for the water height at the inlet (300); -retrieving a new electrical power setpoint value (Pec) greater than the initial value; -transforming the new electrical power setpoint value (Pec) into a characteristic mechanical power (P11); -for said characteristic mechanical power (P11), determining the minimum rotational speed n

Description

CONTROL METHOD FOR A HYDRAULIC POWER UNIT

[001] The invention relates to pumped storage hydroelectric power plants, and in particular the use of such power plants when their turbine drives a generator to supply an AC electrical network.

[002] The hydraulic machines generally include a reversible pump-turbine that is used either in pumping mode or in electrical generation mode. The rotation speed of this reversible pump-turbine should preferably be adapted to the properties of the flow of water passing through it, in order to exhibit an optimal efficiency of use and in order to avoid damaging phenomena such as cavitation.

[003] Most hydraulic machines, when generating electricity to an AC electrical network, rotate at a speed that is synchronous with the frequency of the network to be supplied. For the reversible pumps-turbines in generator mode, this synchronous speed differs generally from the optimum rotation speed defined by the hydraulic properties. The turbines do not therefore generally operate with optimal efficiency in electrical generation mode.

[004] In order to allow an optimal adaptation of the frequency of rotation of the turbine when generating electricity to an AC electrical network, designs of hydroelectric power plants now propose technologies such as DFAM (for double-fed asynchronous machine) or FFSM (for fully-fed synchronous machine). These technologies make it possible to use the turbine in generation mode at a frequency that is different from that of the electrical network to be supplied. The turbine is then used at static operating points with an appropriate rotation speed. The electrical production efficiency and the life of the turbine can thus be enhanced.

[005] With the expansion of the production of electricity by renewable energies, the renewable energy sources connected to the network have potentially intermittent operation, that can lead to temporary and sudden interruptions of the production of electricity. Furthermore, such renewable energy sources lower the inertia of the AC electrical network and therefore its stability. A rapid compensation of the drop in electricity production is therefore essential upon a temporary interruption of one or more renewable energy sources, in order to avoid destabilizing the AC network.

[006] The document entitled "Optimisation des strategies de reglage d'une installation de pompage-turbinage A vitesse variable" [Optimizing strategies for adjusting a variable speed pumping-turbining installation] by Yves Pannatier, proposed using the freewheeling effect of the hydroelectric power plant turbines, to obtain an increase in the electrical power supply to the network with great dynamic range. When the turbine of the hydroelectric power plant is not used at fully load in generation mode, such a freewheeling effect can be used to obtain a very rapid increase in the electrical power generated. The use of freewheeling effect of a turbine consists in making a demand for electrical energy on the generator driven by the turbine, to benefit from the mechanical energy stored in the form of kinetic energy by the rotation speed of the turbine.

[007] The demand for electrical energy induces a drop in the rotation speed of the turbine, made possible by the rotation speed adaptation technologies. The stability of the turbine, in particular a Francis type turbine, can be affected by its rotation speed. Thus, if the rotation speed of the turbine passes below a critical speed, the stability of the turbine can be compromised, and it may then prove impossible to reaccelerate the rotation speed of the turbine. In order to prevent a stability problem, turbine protections disconnect the generator from the electrical network when the problem is detected. The interruption of electrical generation by a hydroelectric power plant that is assumed to apply a rapid electricity production compensation can lead to a cascade effect, culminating in a collapse of the electrical network.

[008] In order to avoid tripping the turbine protections, a minimum rotation speed of the turbine is defined. The increase in electrical energy production is interrupted as soon as the turbine reaches this minimum rotation speed. The minimum rotation speed is determined by a dimensioning of the turbine at a predefined static operating point.

[009] The chosen minimum rotation speed limits the possibilities of temporarily increasing the electrical energy production by the hydroelectric power plant. The increase in electrical power by the hydroelectric power plant can therefore prove insufficient to provide the expected contribution, the sudden interruption of production of one or more renewable energy sources potentially proving to be non-compensated.

[0010] The invention aims to resolve one or more of these drawbacks. The invention thus relates to a method for controlling a hydroelectric power plant comprising a variable speed turbine driving an electrical generator, a conversion structure connected to said generator on one side and to an electrical network on the other side, as defined in the attached claims.

[0011] The invention relates also to the variants of the dependent claims. The person skilled in the art will understand that each of the features of the description or of the dependent claims can be combined independently of the features of a dependent claim, without in any way constituting an intermediate generalization.

[0012] Other features and advantages of the invention will clearly emerge from the description thereof which is given hereinbelow, in an indicative and nonlimiting manner, with reference to the attached drawings, in which: - figure 1 is a diagram illustrating the influence of the reduction of the inertia of an electrical network on its frequency response, as a function of the proportion of renewable energy sources; - figure 2 is a diagram illustrating an example of the trend of electrical power supplied by a hydroelectric power plant in the event of a sudden power compensation; - figure 3 is a diagram illustrating the mechanical power and the speed of a turbine in time, with a power demand inducing an instability of the turbine; - figure 4 is a diagram illustrating a hydroelectric power plant connected to an AC electrical network; - figure 5 is a schematic representation of a diagram illustrating the method implemented to control the hydroelectric power plant; - figure 6 illustrates an example of the trend of the rotation speed of a turbine, compared to thresholds; - figure 7 is a two-dimensional diagram illustrating, on the one hand, the optimum speed of a turbine and, on the other hand, the critical speed of that turbine for different characteristic power values; - figure 8 is an example of hydroelectric power plant control logic; - figure 9 is a diagram illustrating an example of characteristic torque of a turbine as a function of its characteristic speed, for different valve opening values; - figure 10 is a diagram illustrating an example of characteristic flow rate of a turbine as a function of its characteristic speed, for different valve opening values; - figure 11 is a three-dimensional diagram illustrating the hydraulic efficiency of the turbine as a function of its characteristic power and of its characteristic speed; - figure 12 is a diagram illustrating the trend of several operating parameters of a hydroelectric power plant according to the invention; - figure 13 illustrates a diagram of different operating parameters of a hydroelectric power plant of FFSM type; - figures 14 to 16 illustrate simplified control diagrams for a hydroelectric power plant of FFSM type operating with different active power control modes; - figure 17 is a diagram representative of different powers involved in a hydroelectric power plant.

[0013] Figure 1 is a diagram illustrating the influence of the reduction of the inertia of an electrical network on its frequency response, as a function of the proportion of renewable energy sources. The solid line curve corresponds to a reference AC network without renewable energy sources. The dotted line curve corresponds to an AC network having a 20% level of inclusion of renewable energy sources. The dashed line curve corresponds to an AC network having a 40% level of inclusion of renewable energy sources. The chain-dotted line curve corresponds to an AC network having a 60% level of inclusion of renewable energy sources. It can therefore be seen that the increase in the level of inclusion of renewable energy sources lowers the inertia of the AC network and increases the dependency of the network with respect to rapid disconnections of one of the sources, with the risk of destabilizing the network. This risk is quantified by the amplitude of the frequency deviation of the network (which must be limited to 1 Hz for a 50 Hz electrical network for example) and by the rate of frequency variation of the electrical network. The lowering of the frequency of the network can lead to load disconnections, and the rate of frequency variation can activate the load protection and the disconnection of such loads. Such disconnections have a chain effect on the destabilization of the electrical network.

[0014] The invention aims to promote a rapid increase in the electrical power supplied by a hydroelectric power plant. The principle is to make it possible to recover more electrical power by acting on the mechanical inertia of the turbine driving the electrical generator. With a hydroelectric power plant configured to have a variable turbine rotation speed in generation mode, such an inertia effect can be used for the rapid increase in the electrical power, because of the possibility of reducing the rotation speed of that turbine.

[0015] Figure 2 is a diagram illustrating an example of the trend of electrical power supplied by a hydroelectric power plant in the event of a sudden electrical power compensation. At the instant t=0, a demand for increased electrical power is made of the hydroelectric power plant. By virtue of the freewheeling effect, the electrical power supplied by the hydroelectric power plant increases rapidly, in under a second. The new electrical power level supplied by the hydroelectric power plant is maintained for a certain time. The increased electrical power demand is then eliminated, and the electrical power of the hydroelectric power plant reverts to its initial level.

[0016] Figure 3 illustrates an example of mechanical power of the turbine by a solid line, and the speed of that turbine by a dashed line according to an example of failing operation. In this example, an excess of electrical power has been supplied by the hydroelectric power plant, which has caused the turbine to slow down to an excessively low speed. Thus, the turbine enters into a zone of unstable operation, in which the electrical power and the mechanical power drop.

[0017] Figure 4 is a diagram illustrating an example of a hydroelectric power plant 1 connected to an AC electrical network 4 in order to dynamically supply an increase in electrical power without entering into the zone of unstable operation of its turbine. The hydroelectric power plant 1 comprises an electrical circuit 2 and a hydraulic circuit 3. The hydroelectric power plant 1 illustrated here is of FFSM type, but it is also possible to consider other types of hydroelectric power plants, for example of DFIM (for doubly-fed induction machine) type.

[0018] The electrical circuit 2 comprises, as is known per se, a conversion structure on the electrical network side and a conversion structure on the hydraulic machine side that are linked by a direct current link 230. The conversion structure on the electrical network side comprises, as is known, a transformer 250 connected to the electrical network 4, and a DC/AC converter 240 connected to the transformer 250. The conversion structure on the hydraulic circuit side comprises an AC/DC converter 220, an electrical machine 210 connected to the converter 220, and an excitation circuit 200 for the electrical machine 210. The electrical machine 210 is, for example, a motor of synchronous type.

[0019] The hydraulic circuit 3 comprises, as is known, a dam 330 intended to form a water reservoir, a penstock 320 connected to a pipeline entering into the reservoir delimited by the dam 330, a valve, or wicket gate 310 downstream of the penstock 320, and a turbine 300 selectively receiving water based on the state of opening of the valve 310. The turbine 300 is fixed, as is known per se, to the rotor of the electric machine 210. In generator mode, the turbine 300 is driven by water from the penstock 320 and therefore drives the rotor of the electric machine 210 in rotation. In pump mode, the turbine 300 is driven in rotation by the rotor of the electric machine 210 and discharges water into the reservoir via the penstock 320.

[0020] A control circuit (an example of which is detailed hereinbelow) aims to control the conversion structure on the electrical network side, the conversion structure on the hydraulic machine side and the operation of the turbine 300. For a stable operation of the electrical circuit 2, the voltage on the direct current link 230 is maintained constant, by a control of the conversion structure on the electrical network side. The control structure on the hydraulic circuit side 3 and the turbine 300 are controlled by this control circuit in order to deliver the target electrical power.

[0021] To this end, the control circuit can have a first mode, called active electrical power control mode, controlling the active electrical power delivered by the converter 220 to the electrical network 4, and by controlling the rotation speed of the turbine 300 by a setpoint value on the level of opening of the wicket gate 310.

[0022] In this mode of operation, when the control circuit applies an increased setpoint power to the converter 220, the electrical power delivered increases effectively after a very short time (for example of the order of 20 ms), the rotation speed of the turbine 300 being slowed down. Without instability, the control circuit manages to control the turbine 300 in order for the latter to revert to its initial rotation speed.

[0023] The control circuit can have a second mode, called angular speed control mode, in which the rotation speed of the turbine 300 is controlled via the converter 220, while the active power supplied to the network 4 is controlled via the turbine 300.

[0024] In this mode of operation, the control circuit controls the rotation speed of the turbine 300 for this speed to supply an optimum efficiency for the new increased setpoint power. The control circuit then controls the converter 220 so that the speed is adjusted by adapting the output power to the electrical network 4 as a function of the mechanical power supplied by the turbine 300.

[0025] For the example of continuous use of a turbine 300 of Francis type, the continuous operation range is usually defined by constraints of cavitation at the input, cavitation at the output, of inter-blade vortex of the turbine 300, of full-load vortex or of partial-load vortex.

[0026] For a fixed speed reversible pump-turbine, the limiting factor for a dynamic response of the turbine 300 is the water pressure variation in the penstock 320 connected to the turbine 300. The pressure variation due to the inertia of the water column is then an indicator of the hydroelectric stability of the turbine 300. For a turbine 300 operating at variable speed, it is necessary to incorporate the speed limits in the event of a power dynamic response because of the limits of the freewheeling effect.

[0027] The example of a turbine 300 of Francis type is detailed here, but other types of turbines can also be implemented in the context of the invention, for example a turbine of Kaplan or VLH type.

[0028] The control of the hydroelectric power plant 1 implements a method that is illustrated schematically in figure 5 in order to determine a minimum rotation speed of the turbine 300 that guarantees its stability. This method comprises a set 10 of design steps, and a set 11 of operation control steps.

[0029] The modeling of the dynamic behavior of the hydraulic machine or of the hydraulic circuit 3 is based on measurements of operating points for different levels of opening of the wicket gate 310 of the turbine 300, in static operation. The modeling of the dynamic behavior of the hydraulic circuit 3 is based on the assumption that the operating points of the hydraulic circuit 3 in dynamic mode operation are identical to its operating points in steady state operation.

[0030] Steady-state modelings of operation of hydraulic machines generally define their performance based on 5 variables, generally expressed in units [pu]: - the level of opening of the wicket gate 310 of the turbine 300, designated by the

parameter g; - the torque of the turbine 300, designated by the parameter T; - the water flow rate through the turbine 300, designated by the parameter q; - the angular rotation speed of the turbine 300, designated by the parameter n; - the water head in the penstock 320, designated by the parameter h.

[0031] The recovery of the information relating to the steady-state operating points of the hydraulic circuit 3 corresponds to the step 101 of the design steps.

[0032] In order to eliminate the water head parameter, parameter transformations are performed to obtain characteristic parameters, independent of the water head h: Ti1=T/h n11=n/4h q11=q/4h

[0033] The modeling of the operation of the hydraulic circuit 3 can then be defined relative to the variables g, Ti, nil, qi. The modeling can then be illustrated on the one hand by a diagram of characteristic torque Ti as a function of the characteristic speed nil for different opening values g (example of figure 9), and on the other hand by a diagram of characteristic water flow rate q11 as a function of the characteristic speed nil for different opening values g (example of figure 10). The diagrams of figures 9 and 10 correspond to an example of Francis type turbine.

[0034] It is then possible to deduce therefrom a characteristic mechanical power P11: Pii= Ti * ni= P/hih

[0035] The efficiency of the turbine can then be expressed by q= P1i/qii.

[0036] Corresponding diagrams can thus be generated for the parameters r and P11.

[0037] The transformations of parameters and the generation of modeling diagrams of characteristic parameters as a function of the characteristic rotation speed nil can be performed in step 102 of the design steps.

[0038] The diagrams of figures 9 and 10 can be used to deduce a three-dimensional diagram of the efficiency of the hydraulic circuit 3 as a function of the characteristic mechanical power P11 and of the characteristic rotation speed nil. The corresponding diagram is illustrated in figure 11. The generation of such a diagram corresponds to the step 103 of the design steps. The right edge of the three-dimensional diagram here illustrates the critical rotation speeds of the turbine 300 as a function of the characteristic power P11. The thick line curve corresponds to the optimal operating points of the turbine 300 as a function of the characteristic power P11. The (solid line) curve of optimal operating speed of the turbine 300 and the (dashed line) curve of the critical speed of this turbine 300 are illustrated in the two-dimensional diagram of figure 7.

[0039] In the case of this invention, the critical speed can be defined as the minimum speed for which the turbine 300 can keep the mechanical power equivalent to the electrical power demanded by the electrical network 4. Thus, beyond this critical speed, the hydraulic machine cannot maintain the electrical power demanded the network and will enter into a cascade effect leading to an instability.

[0040] An example of set of controls 11 of the hydraulic circuit 3 for the implementation of the invention will now be described.

[0041] In the step 117, the water head value h applied to the input of the turbine 300 is recovered. This water head value h can be recovered by any appropriate means, for example by pressure sensors as is known per se. The water head h is a parameter that is relatively easy to recover, and that varies at a very slow speed, with the water level upstream of the dam 330.

[0042] In the step 116, a new setpoint mechanical power value Ph is recovered, in order to cope with a demand for dynamic increase in electrical power to be supplied to the network 4. The mechanical power setpoint corresponds to the mechanical power of the turbine 300 necessary to supply a desired electrical power Pes to the network 4. The mechanical power setpoint Pch can for example be determined from an electrical power setpoint Pec: - by applying a transformation taking in account the efficiency of conversion between the mechanical power and the electrical power supplied to the network, in steady state operation; - by taking into account the following relationship in dynamic operation:

Pelec = q*Pmeca + d(Ht * n 2)/dt, with ri the conversion efficiency between the mechanical power and the electrical power supplied to the network, Ht the inertia of the rotationally driven mass of the hydraulic circuit 3 (combination of the turbine 300 and of the rotor of the electric motor 210) expressed in seconds, and n the rotation speed of the turbine 300 at the nominal electrical power.

[0043] The setpoint electrical power Pec can for example correspond to the following relationship: Pec = Pei + Pea with Pei the initial electrical power supply to the network 4, and Pea an electrical power increase requested by the electrical network 4.

[0044] Figure 17 schematically details the different powers involved in the hydroelectric power plant 1. The downward arrows correspond to the power losses, the upward arrows correspond to the temporarily-exchanged energies.

[0045] The dam 330 has a potential energy Epot.

[0046] In the penstock 320, a hydraulic power Phyd passes through, there are friction losses Pfri and an energy AEWis temporarily exchanged.

[0047] At the valve 300 and turbine 310 assembly, a mechanical power Pmec is developed. A power Pt is lost through hydraulic losses in this assembly, and an energy AEp- is temporarily exchanged.

[0048] At the electrical machine 210, an electrical power Pelec is supplied. A power Plm is lost and an energy AERI is temporarily exchanged because of the rotational inertia of the rotating assembly.

[0049] At the direct current link 230, a power Ptrans is transmitted. A power Plc is lost and an energy AEc is temporarily exchanged through capacitive effect.

[0050] At the transformer 250 and the electrical network 4, a power Pout is supplied. A power Pltr is lost through transformation losses.

[0051] In step 115, a conversion of the setpoint mechanical power Pch into a characteristic power P11 is performed by taking account of the water head h by means of the formula P11=Ph/h.

[0052] In the step 111, the critical characteristic rotation speed n11era of the turbine 300 is determined for this characteristic power P11, for example from the diagram of figure 7 or the diagram of figure 11. This critical characteristic rotation speed nicrit is converted in step 112 into critical rotation speed ncrt of the turbine 300, using the relationship ni=n/h.

[0053] Advantageously, the optimal characteristic rotation speed niopt of the turbine 300 for this characteristic power P11 in the stabilized state is also determined in step 111. This optimal characteristic rotation speed n1lopt is converted in the step 113 into optimal rotation speed nopt of the turbine 300.

[0054] In the step 114, a regulation of the hydraulic circuit 3 is performed so as to guarantee that the rotation speed of the turbine 300 remains greater than the value nert in the event of a dynamic power increase supplied by the hydroelectric power plant 1. Thus, it is possible to guarantee that the rotation speed of the turbine 300 is maintained at a level that avoids the tripping of instability detection safety devices, which makes it possible to avoid a disconnection of the hydroelectric power plant 1 from the network 4.

[0055] Advantageously, the regulation of the power plant 1 can be performed so as to guarantee that the rotation speed of the turbine 300 remains greater than a value ns, with ns > norit (for example ns = 1.05 * not), so as to retain a safety margin relative to the critical rotation speed nerit.

[0056] In the step 114, the regulation of the hydraulic circuit 3 can also be performed to operate the latter at its optimal efficiency after the mechanical power has reached the mechanical power Pch value. In fact, once this mechanical power is established, it is advantageous to stabilize the hydraulic circuit 3 for it to operate at its optimal efficiency.

[0057] The values nopt and nerit can be supplied to the control circuit of the hydroelectric power plant 1 for the real-time control thereof.

[0058] Upon a request to increase electrical power received from the network 4, the method for controlling the hydroelectric power plant can be implemented according to the following loop: a) receive the desired electrical power Pes; b) recover the water head value at the input of the turbine 300; c) calculate the critical rotation speed norit of the turbine 300 for the power Pes; d) determine the safety rotation speed ns of the turbine 300; e) recover the value of the rotation speed n of the turbine 300; f) determine whether n 2ns; g) if n 2ns, define Pec=Pes; h) if n< ns, define Pec =Pes * (n- norit)/(ns - nerit); i) return to the step a)

[0059] Such a method makes it possible to guarantee that the turbine 300 will gradually revert to a rotation speed greater than the value ns.

[0060] If mechanical power reasoning is used, the values Pes and Pec can be replaced respectively by Ph and Pminter respectively, with Pminter an intermediate mechanical power setpoint.

[0061] Figure 6 illustrates an example of trend of the speed of a turbine 300 in the event of a dynamic increase in the electrical power supplied by the hydroelectric power plant. At the instant t=0, an increased power demand is received by the hydroelectric power plant. The hydroelectric power plant 1 supplies the increased electrical power to the network 4 within a very short time. The rotation speed of the turbine 300 decreases gradually, through the braking by the motor 210 used as generator to supply the increased electrical power. When the rotation speed of the turbine 300 reaches the value ns, the control circuit of the hydroelectric power plant reacts to avoid having the rotation speed fall below the threshold nrt. The control circuit for example slightly reduces the electrical power supplied to the network 4 or by an increased opening of the valve 310.

[0062] According to one desirable mode of operation, it is possible to anticipate beforehand the quantity of kinetic energy available from the turbine 300 without destabilizing it, in order to determine a transient electrical power which can be demanded by the electrical network 4 and for what period of time that transient electrical power can be demanded.

[0063] Thus, the method described previously made it possible to calculate the values nopt and ncrit. The quantity of kinetic energy available Wec can be defined by the following relationship: Wec = Ht * (noptinit 2 - ncrit2 )

with Ht the inertia of the rotationally driven mass of the hydraulic circuit 3 (combination of the turbine 300 and of the rotor of the electric motor 210), expressed in seconds, and noptinit the optimal rotation speed of the turbine 300 used before the additional power demand.

[0064] The inertia Ht can be determined by the following formula, for a nominal rotation speed: Ht= J* nnom2 / (2*Sn) with J the moment of inertia of the rotationally-driven mass in Kg.m 2 , nnom the nominal speed (for example for an operating point with optimal efficiency) in rad/s, Sn the nominal power of the electric motor in VA.

[0065] Simulations have for example been carried out for an hydraulic circuit 3 exhibiting an inertia of 3.2 seconds for the nominal water head of the hydroelectric power plant.

[0066] If the electrical network 4 requires an increase Wreq of energy supplied that is less than the quantity of kinetic energy available Wec, the hydroelectric power plant can supply this energy.

[0067] The energy increase Wreq is defined as follows. Initially, the hydroelectric power plant 1 observes the condition Pmecainit * R= Pelecinit, with Pmecainit the initial mechanical power of the turbine 300, R the efficiency of conversion of the mechanical energy of the turbine 300 into electrical power supplied to the network 4, and Pelecinit the initial electrical power supplied to the network 4. When the network 4 demands an increased electrical power Pec = Pch * R, the electrical power increases very rapidly to the value Pec. According to one model, it is possible to estimate that the mechanical power increase of the turbine 300 is linear in the time between the value Pmecaint and Pob. Thus, a duration tp is necessary before the mechanical power reaches the value Poh, which can be reflected by a slope designated by RoCoP. The value Wreq can then be defined by the following relationship: Wreq = (Pch- Pmecainit) 2 /(2*RoCoP)

[0068] Thus, it will be possible to determine beforehand whether the hydroelectric power plant 1 is capable of supplying the energy Wreq without inducing instability of the hydraulic circuit 3. It will therefore be possible beforehand to calculate the maximum value of Pec or of Ph available when the hydroelectric power plant 1 operates at a given operating point.

[0069] If it is determined that Wreq > Wee, it is possible to calculate a mechanical power value Plim that satisfies Plim <Ph and (Pim-Pmecainit)2 /(2*RoCoP) < Wee. This value Plim can be used temporarily as setpoint for the hydroelectric power plant 1 to increase the electrical power supplied to the network 4 by freewheeling effect.

[0070] RoCoP is here approximated as a fixed slope (test results have shown that this assumption was fairly accurate) but can in practice be a variable parameter dependent on the efficiency of the turbine, on the inertia of the water column, on the water hammer effect and on the response time of the valve 310. RoCoP can be expressed as the maximum rate of mechanical power variation Pm of the turbine 300, i.e. dPm/dt.

[0071] A corresponding control method is illustrated with reference to figure 8. The control method here implements an example of set of controls 12 of the hydraulic circuit 3. The control steps 111 to 113 and 115 to 117 are, here, identical to those described previously.

[0072] In the step 118, the inertia value Ht described previously is recovered.

[0073] In the step 119, the quantity of kinetic energy Wee available through freewheeling effect is calculated, according to the relationship described previously, on the basis of the values Ht, noptinit and nrit available. This calculation makes it possible for example to inform the operator of the hydroelectric power plant 1 or the operator of the network 4 as to the value of this quantity of energy We, to anticipate the management of a capacity to increase electrical power produced by the hydroelectric power plant 1.

[0074] In the step 120, the condition Wreq > Wee is checked, as detailed previously. If this is not the case, Pch can be used as mechanical power setpoint in the step 122. If Wreq > Wee, the step 121 is implemented: the mechanical power value Plim that satisfies Plim <Pch and (Plim-Pmecainit) 2 /(2*RoCoP) < Wee is calculated. Then, the hydroelectric power plant 1 is temporarily controlled with this mechanical power setpoint value Plim.

[0075] The invention makes it possible to envisage new economic models for operating a hydroelectric power plant. In practice, an operater of hydroelectric power plants can operate the latter at low load and market a dynamic electrical power reserve to the operator of the electrical network. In fact, the operator of the electrical network may need to pay for this relatively expensive power reserve, in order to compensate for any interruptions of service from renewable energy sources such an economic model is all the most cost effective since the invention makes it possible to accurately determine the power reserve, which makes it possible to market a power reserve of an increased amount.

[0076] The manager of the electrical network 4 to which the hydroelectric power plant is connected can anticipate the management of its network, by potentially having available, in real time, the power reserve from the hydroelectric power plant 1.

[0077] Figure 12 is a diagram illustrating the trend of several operating parameters of a hydroelectric power plant 1 in time, according to an example of implementation of the invention. The continuous line curve corresponds to the electrical power supplied to the network 4, the dotted line curve corresponds to the mechanical power of the hydroelectric power plant 1, the broken line curve corresponds to the rotational speed of the turbine 300, the chain-dotted limit corresponds to the safety rotation speed ns of the turbine 300, the double-dot-dash limit corresponds to the critical rotation speed norit.

[0078] At t=0, the mechanical power and the electrical power of the hydroelectric power plant 1 are at a level of 0.3 in [pu], i.e. 30% of their nominal value. A demand for an electrical power of 0.9 [pu] is received for the network 4 at t=0. Through control of the conversion structure, the electrical power of 0.9 [pu] is supplied in a time of the order of 0.3s. The turbine 300 is controlled with a power setpoint Pch, corresponding to the electrical power of 0.9 [pu]. A corresponding opening level is applied to the valve 310.

[0079] Between the instant t=0.3s and t=1.2s, most of the increase in electrical power is supplied by the freewheeling effect of the turbine 300. The rotation speed of the turbine 300 therefore decreases progressively and the mechanical power increases progressively within this time interval. The electrical power is maintained at a level of 0.9 [pu].

[0080] Between the instant t=1.2s and t=2.3s, the electrical power is maintained at the level of 0.9 [pu]. The mechanical power increases more slowly over this interval, the increase in the mechanical power then being due mainly to the increase in the level of opening of the valve 310. The rotation speed of the turbine 300 continues to decrease, while remaining greater than the value ns.

[0081] At the instant t=2.3s, the rotation speed of the turbine 300 reaches the value ns. Between the instant t=2.3s s and t=2.6s, a limitation of the electrical power is requested, as described when the condition n<ns is fulfilled. The electrical power decreases, while the rate of increase in the mechanical power increases. The rotation speed of the turbine 300 decreases but remains greater than the value nerit. At the instant t=2.6s, the mechanical power reaches the electrical power value, approximately at 0.61 [pu].

[0082] Between the instant t=2.6s s and t=3.5s, the mechanical power of the turbine 300 and the electrical power increase, to reach the value of 0.9 [pu]. The rotation speed of the turbine 300 increases to exceed the value ns.

[0083] After t=3.5s, the mechanical power exceeds 0.9 [pu], such that the rotation speed of the turbine 300 continues to increase. The turbine 300 is controlled to maintain a mechanical power greater than 0.9 [pu] such that its rotation speed continues to increase to n= nopt.

[0084] The calculation of the critical rotation speed of the turbine 300 is based here on the hydraulic parameters of the hydraulic circuit 3. Other constraints of electrical nature can also limit the increased electrical power that the hydroelectric power plant 1 can supply. Current or flux limitations in the electrical circuit 2 can thus also limit the increase in electrical power that the hydroelectric power plant 1 can supply dynamically. Such electrical limits are generally defined in the design of the hydroelectric power plant 1 and more often than not do not require calculation in real time.

[0085] Figure 13 illustrates a simplified control diagram for a hydroelectric power plant 1 of FFSM type.

[0086] The different operating variables of the hydroelectric power plant 1 are, here: -Vrd, Vrq: respectively, the active voltage and the reactive voltage on the transformer 250; - V,V: respectively the active voltage setpoint and the reactive voltage setpoint on the converter 240; -Vd,V': respectively the active voltage setpoint and the reactive voltage setpoint on the converter 220; -VD: the voltage on the link 230; -Vexc: the voltage on the excitation circuit 200; -g the level of opening of the wicket gate; -Id, lq: respectively the active current and the reactive current between the converter 240 and the transformer 250; -Id', lq': respectively the active current and the reactive current between the converter 220 and the electrical machine 210; -If, the armature current; -w, the rotation speed of the electrical machine 210; -Qs, the reactive power of the electrical machine.

[0087] The degrees of freedom of the control of the hydroelectric power plant 1 are: -Vd,Vq; -Vd',Vq'; -g; -Vexe.

[0088] These degrees of freedom make it possible to control the objectives for the following output metrics: -VDC; -Vtrd, Vtrq; -the reactive power of the electrical machine; -If or the electromagnetic flux proportional to If; -the rotation speed of the electrical machine; -Vs the stator voltage of the electrical machine.

[0089] The table below reviews the influence of each of the degrees of freedom on output metrics:

Degree of freedom Possible output metrics Vd Id, VDC Vq ,1q, Vtrd, Vrq Vd' Active power of the machine (torque, Id'), VDC, W Vq' Reactive power of the machine (Vs, Iq') g Torque of the electrical machine, w Vexc Electromagnetic flux (reactive power of the machine, Vs, If)

[0090] Among the global variables of the hydroelectric power plant 1, the following can be identified: -the output active power: necessary for the power system; -the voltage VDC: necessary for the operation of the converters, indicative of the active power balancing between the converter 220 and the converter 240; -the rotation speed w: necessary for an optimal efficiency, indicative of the active power balancing between the electrical machine and the hydraulic system.

[0091] Among the local variables that make it possible to optimize operation, the following can be identified: -the output reactive power: impacts the operation of the power system. Influences only the converter 240; -the reactive power of the machine: impacts the optimum operation of the machine. Influences only the converter 220 and the excitation circuit 200; -electromagnetic flux of the machine: impacts the optimum operation of the machine. Must be controlled to avoid a saturation of the magnetic circuit or an overvoltage at the terminals of the machine. Influences only the excitation system and the converter 220.

[0092] The following control modes of the hydroelectric power plant 1 can for example be envisaged:

-an active power control mode. Such a control mode makes it possible to have a rapid response in the active power, of the order of ten or so milliseconds. The objective is to control the output active power of the hydroelectric power plant 1. The control variables used are then, in order of priority, the active voltage Vd' and the active voltage Vd and the level of opening g; -the voltage VDC control mode. The objective is to control the voltage VDc. The control variables used are then, in order of priority, the active voltage Vd and the active voltage Vd' and the level of opening g; -angular speed w control mode. The control variables used are then, in order of priority, the level of opening g, the active voltage Vd' and the active voltage Vd.

[0093] Figure 14 illustrates more specifically the theoretical diagram of control of the hydroelectric power plant in active power control mode.

[0094] The reference 301 designates a control and hydraulic calculation module. The reference 311 designates a control circuit for the valve associated with the turbine 300 (for example a servomotor of this valve). The reference 201 designates an excitation control module. The reference 202 designates a flux limiter. The reference 221 designates a control circuit of the converter 220. The reference 241 designates a control circuit of the converter 240.

[0095] The control module 301 receives the water head h and the power setpoint Pre as input parameters. The control module 301 determines a rotation speed setpoint Wrefand supplies it to the control circuit 311. The control circuit 311 supplies an opening level control setpoint g and receives the value of the rotation speed w of the turbine.

[0096] The flux limiter 202 receives the value w from the turbine 300 and supplies a setpoint VSREF to the excitation control module 201. The module 201 receives a voltage value Vs and applies an excitation voltage setpoint Vexc to the excitation circuit 200.

[0097] The circuit 221 receives the value Pref, the power P (real electrical power on the network 4 or real power of the machine 210) supplied to the network 4 and the reactive power value QSM of the machine 210. The circuit 221 applies the voltage setpoints Vd' and Vq' to the converter 220. The circuit 241 receives the value Qref, the voltage VDc, and the reactive power Qtr between the converter 240 and the transformer 250. The circuit 241 applies the voltage setpoints Vd and Vq to the converter 240.

[0098] For example, if the hydroelectric power plant 1 receives a command Pref to change active power of the network 4, the converter 220 adapts the output power of the electrical machine 210 to the new power reference. This adaptation can typically be implemented within a delay of 50 ms after the command to change active power.

[0099] The converter 240 then adapts the active power extracted to regulate the voltage VDC. The regulation of the voltage VDC can for example be effective within a delay of the order of 200 milliseconds after the command to change active power.

[00100] The optimal rotation speed of the turbine 300 is calculated by means of the module 301, for the requested output power. The calculation can be done within a delay of a few tens of milliseconds after the command to change active power. The rotation speed of the turbine 300 is then adjusted by the valve, for example to increase the mechanical power supplied by the hydraulics, in order to adjust the rotation speed of the turbine 300 to its optimal value. The adjustment of the rotation speed and the use of the freewheeling effect can typically be continued for a period of the order of 120 seconds after the command to change active power.

[00101] Figure 15 more specifically illustrates the diagram of control of the hydroelectric power plant in control mode based on the direct voltage VDC. The components illustrated here are identical to those described with reference to figure 14.

[00102] The control module 301 receives the water head h and the power setpoint Pref as input parameters. The control module 301 determines a rotation speed setpoint Wrefand supplies it to the control circuit 311. The control circuit 311 supplies an opening level g control setpoint and receives the value of the rotation speed w of the turbine.

[00103] The flux limiter 202 receives the value w from the turbine 300 and supplies a setpoint VSREF tothe excitation control module 201. The module 201 receives a voltage value Vs and applies an excitation voltage setpoint Vexc to the excitation circuit 200.

[00104] The circuit 221 receives the value Pref, the voltage VDc, and the reactive power value QsM of the machine 210. The circuit 221 applies the voltage setpoint V' and Vq' to the converter 220. The circuit 241 receives the value Qref, the power P, and the reactive power Qtr between the converter 240 and the transformer 250. The circuit 241 applies the voltage setpoints Vd and Vq to the converter 240.

[00105] For example, if the hydroelectric power plant 1 receives a command Pref to change active power from the network 4, the converter 240 adapts the output power to the new power reference. This adaptation can typically be implemented with a delay of 50 ms after the command to change active power.

[00106] The converter 220 then adapts the active power of the machine 210 to regulate the voltage VDc. The regulation of the voltage VDcCan for example be effective within delay of the order of a 200 milliseconds after the command to change active power.

[00107] The optimal rotation speed of the turbine 300 is calculated by means of the module 301, for the requested output power. The calculation can be performed within a delay of a few tens of milliseconds after the command to change active power. The rotation speed of the turbine 300 is then adjusted by the valve, for example to increase the mechanical power supplied by the hydraulics, in order to adjust the rotation speed of the turbine 300 to its optimal value. The adjustment of the rotation speed and the use of the freewheeling effect can typically be continued for a period of the order of 120 seconds after the command to change active power.

[00108] Figure 16 more specifically illustrates the theoretical diagram of control of the hydroelectric power plant in control mode based on the angular speed w. The components illustrated here are identical to those described with reference to figure 14.

[00109] The control module 301 receives the water head h and the power setpoint Pref as input parameters. The control module 301 determines a rotation speed setpoint Wrefand supplies it to the control circuit 311. The control circuit 311 supplies an opening level g control setpoint and receives the value of the rotation speed w of the turbine 300.

[00110] The flux limiter 202 receives the value w from the turbine 300 and supplies a setpoint VSREF tothe excitation control module 201. The module 201 receives a voltage value Vs and applies an excitation voltage setpoint Vexc to the excitation circuit 200.

[00111] The circuit 221 receives the value w, the valuewref, and the reactive power value QsM of the machine 210. The circuit 221 applies the voltage setpoints Vd' and Vq' to the converter 220. The circuit 241 receives the value Qre, the voltage VDC, and the reactive power Qtr between the converter 240 and the transformer 250. The circuit 241 applies the voltage setpoints Vd and Vq to the converter 240.

[00112] For example, if the hydroelectric power plant 1 receives a command Pref to change active power from the network 4, the optimal rotation speed of the turbine 300 is calculated by means of the module 301, for the requested output power. The calculation can be done within a delay of a few tens of milliseconds after the command to change active power.

[00113] The mechanical power of the machine 300 is then adjusted by the valve, for example to increase the mechanical power supplied by the hydraulics, in order to converge toward the power setpoint requested by the network 4.

[00114] The converter 220 then adapts the rotation speed of the turbine 300 for the real rotation speed to converge toward the calculated optimal rotation speed.

[00115] Throughout this process, the converter 240 adapts the electrical power transferred to the network 4, in order to regulate the voltage VDC to its nominal value.

[00116] In the case of a drop in frequency on the AC network 4, for example due to a loss of a source which is connected to it, the invention can advantageously make it possible to contribute to the compensation for the corresponding drop in inertia. To this end, the hydroelectric power plant 1 can be driven in order to emulate an inertia on the latter that is greater than its real inertia. To this end, the rotation speed of the turbine 300 is modified proportionally to the frequency variation on the network 4, according to the following relationship: An = Ksynth * Af with An the controlled rotation speed variation, Af the transient frequency variation value on the network 4, and Ksynth an emulated inertia constant. By virtue of the properties of speed variation for the turbine 300, it is possible to emulate an inertia of the rotating mass at the value Ksynth, greater than its real inertia. Provided that the rotation speed of the rotating mass is maintained at a value greater than the critical speed ncrit, it is thus possible to reduce the rotation speed of the turbine 300 in order to contribute to the compensation for the drop in inertia of the network 4.

[00117] The maximum value Ksm of the emulated inertia constant Ksynth can be defined as follows: Ksm= (noptinit - ncrit)/ Afmax with Afmax the lower limit of the maximum transient variation of frequency as specified by the operator of the network 4.

[00118] In the example described previously, the network 4 is of alternating current type. It is also possible to provide for connecting the hydroelectric power plant 1 to a direct current network 4, with an appropriate conversion structure.

[00119] In the examples described previously, the hydraulic machine 300 is operated in turbine mode. It is also possible to provide the same control modes for the hydroelectric power plant 1 operated in pump mode.

Claims

CLAIM

1. A method for controlling a hydroelectric power plant (1) comprising a variable speed turbine (300) driving an electrical generator (210), a conversion structure (220, 230, 240) connected to said generator (210) on one side and to an electrical network (4) on the other side, comprising the steps of: - using a law of minimum characteristic rotation speed as a function of a characteristic mechanical power delivered by the turbine, this minimum characteristic rotation speed guaranteeing the stability of the turbine; - operating the hydroelectric power plant with an initial setpoint value of electrical power to be supplied to the electrical network; - recovering a water head value at the input of the turbine (300); - defining a new electrical power setpoint value (Pec) greater than the initial value; - transforming the new electrical power setpoint value (Pec) into characteristic mechanical power (P11) to be delivered; - based on said law, for said characteristic mechanical power to be delivered (P11), determining the minimum rotation speed ncrit of the turbine (300); - controlling the hydroelectric power plant (1) to deliver an electrical power equal to the new electrical power setpoint value (Pec) and to maintain the rotation speed of the turbine (300) above the determined minimum rotation speed nerit.

2. The control method as claimed in claim 1, wherein said step of controlling the hydroelectric power plant (1) to deliver an electrical power equal to the new electrical power setpoint value (Pec) comprises the application of successive electrical power setpoints increasing between the initial electrical power setpoint value and said new electrical power setpoint value (Pec).

3. The control method as claimed in claim 2, comprising: - transforming the successive electrical power setpoints into successive characteristic mechanical power setpoints; - based on said law, for each of the characteristic mechanical power setpoints, determining the corresponding minimum rotation speed of the turbine (300).

4. The control method as claimed in any one of the preceding claims, the conversion structure being configured to convert an alternating voltage generated by the generator into an alternating voltage exhibiting a frequency that is different from that generated by the generator or into a direct voltage, the method also comprising the steps of: - receiving a new electrical power value (Pes) to be supplied to the electrical network; - converting said new electrical power value (Pes) to be supplied into a new mechanical power setpoint value (Ph); - transforming the new mechanical power setpoint value Poh into said characteristic mechanical power (P11) to be delivered.

5. The control method as claimed in any one of the preceding claims, wherein the hydroelectric power plant (1) is controlled to deliver an electrical power Pes to the electrical network that is equal to the new electrical power setpoint value (Pec) and to maintain the rotation speed of the turbine (300) above a safety rotation speed ns, with ns > norit.

6. The control method as claimed in claim 5, comprising the steps of: - a) recovering the instantaneous rotation speed n of the turbine (300); - b) if ns > n > norit, controlling the hydroelectric power plant (1) with an electrical power setpoint Pec = Pes* (n- norit)/ns - norit); - c) repeating the steps a) and b).

7. The control method as claimed in any one of the preceding claims, comprising the steps of: - recovering the maximum speed RoCoP of variation of mechanical power Pm of the turbine (300), with RoCoP=dPm/dt; - recovering the inertia Ht of the mass including the turbine (300) and the generator (220); - calculating a quantity of kinetic energy We available through freewheeling effect, with Wec= Ht * (noptinit 2 - ncrit 2 ), with noptinit the speed for the initial setpoint value of electrical power to be supplied to the electrical network.

8. The control method as claimed in claims 4 and 7, comprising the steps of: - determining an initial mechanical power setpoint value Pmecainit corresponding to the initial electrical power value to be supplied to the electrical network; - if Wreq > Wec, with Wreq = (Pch-Pmecainit) 2 /2*RoCoP), calculating a mechanical power value Plim that satisfies Plim <Pch and (Pim-Pmecainit)2 /(2*RoCoP) < Wec; - temporarily controlling the hydroelectric power plant (1) with a mechanical power setpoint value equal to Plim or with an electrical power setpoint value to be supplied to the electrical network corresponding to this mechanical power setpoint Plim.

9. The control method as claimed in any one of the preceding claims, wherein said controlled hydroelectric power plant (1) includes a conversion structure (220, 230,

240) of FFSM type including a first AC/DC converter (220) connected to said electric motor (210).

10. The control method as claimed in claims 4 and 9, wherein said hydroelectric power plant (1) is controlled in active power control mode by applying the new electrical power value (Pes) to be supplied to the electrical network to a control circuit of the first AC/DC converter (220).

11. The control method as claimed in claims 4 and 9, wherein said hydroelectric power plant (1) is controlled in control mode based on a direct voltage in the conversion structure, by applying a setpoint for said direct voltage to a control circuit of the first AC/DC converter (220).

12. The control method as claimed in claims 4 and 9, wherein said hydroelectric power plant (1) is controlled in control mode based on the rotation speed of the turbine, by applying a setpoint for said rotation speed of the turbine to a control circuit of the first AC/DC converter (220).

13. The control method as claimed in any one of the preceding claims, wherein: - a drop in frequency is determined on said electrical network (4); - the new electrical power set point value (Pec) is defined so that the rotation speed of the turbine (300) declines proportionally to the determined decrease in frequency.