Process for controlling isolated microgrids based on power-based control and modified Volt-VAr function methods

Abstract

A control process for microgrids for voltage regulation on the main bus and power factor (PF) regulation at generator terminals is presented, especially in events scheduled in the microgrid that result in electrical transients, such as direct starting of induction motors (IM). The technology takes advantage of idle capacity of distributed converters (for example: frequency inverters, variable frequency drive or VFD) of microgrids making them, in coordinated manner, injecting and/or absorbing reactive power, in addition to exploit the reduced latency of autonomous VFD control during the transient. The Power-Based Control (PBC) technique is used and a modified Volt-VAr function is created applied during the transitional regime.

Claims

1. A process for controlling isolated microgrids based on Power-Based Control (PBC) method and modified Volt-Var function method, the process comprising: a) identifying, for a microgrid to be controlled, a range of power factor (PF) values between a minimum value (PF*min) and a maximum value (PF*max); b) in steady state: applying the PBC method via a central controller (CC); and defining PF references to be used for generators for the maximum value (PF*max) to control frequency inverters to inject a reactive power calculated via the CC using the PBC method; c) before a transitional regime starts: changing the PF references to be used for the generators to the minimum value (PF*min); and programming the frequency inverters to use a unitary PF as a reference such that the frequency inverters include an increase in availability of reactive power exchange with the microgrid within a technically specified limit of the frequency inverters; d) during the transitional regime: suspend the CC from applying the PBC method; and start the frequency inverters to define the reactive power exchanged with the microgrid based at least in part on a voltage measured at a connection point of the frequency inverters using a modified Volt-Var curve; e) after the transitional regime: resuming the PBC method via the CC; returning the PF references to be used for the generators to the maximum value (PF*max); and injecting, by the frequency inverters, the calculated reactive power.

2. The process of claim 1, wherein starting the frequency inverters includes autonomously starting the frequency inverters.

3. The process of claim 1, wherein suspending the CC and starting the frequency inverters includes suspending the CC and starting the frequency inverters from a start of the transitional regime until the transitional regime ends.

Description

BRIEF DESCRIPTION OF FIGURES

(1) FIG. 1 shows a non-limiting example of an isolated microgrid in which the proposed process was applied through a computer simulation. The microgrid represents a microgrid of an oil platform, in a non-limiting way, and is formed by the generators G01, G02, G03, G04, whose respective keys (C1, C2, C3 and C4) connect them to the main bus (01) respectively in the connection points A03, A02, B02 and B03; and a bus sectioning circuit breaker (Bus-Tie). The Power Management System (PMS) unit is the central controller and receives measurements of voltage and current on the main bus (ug and ig) and receives information from power (output reactive power, Qi,ufd, and maximum reactive power available, Qi,max) of the i.sup.th frequency inverter distributed in the electrical grid via a low-bandwidth communication bus. The bus (02) is connected to the main bus (01) and receives connections from the following devices: an aggregate motor (M1), a transformer (T1) connected to a low voltage load (L1), a medium voltage load (L2), a motor (M2) that can be directly connected to the bus (02) through the key (C5), a motor (M3) which can be connected to the bus (02) through a VFD converter (I1) that receives information from the PMS containing the value of the coefficient b to be used in calculating the reactive power that must be exchanged with the microgrid and transmits power information (output reactive power, Qi,ufd, and maximum reactive power available, Qi,max, i=1 for this inverter); a transformer (T2) connected to a low voltage load (L3), a circuit breaker Dj1, a circuit breaker Dj3, a transformer (T3) connected to a low voltage load (L4), a motor (M4) which can be connected to the bus (02) through a VFD converter (I2) which receives information from the PMS containing the value of the coefficient b to be used in calculating the reactive power that must be exchanged with the microgrid and transmits power information (output reactive power Qi,ufd, and maximum reactive power available, Qi,max, i=2 for this inverter), and an aggregate motor (M5). The bus (02) is connected to the bus (03) which receives connections from the following devices: an aggregate motor (M6), a transformer (T4) connected to a low voltage load (L5), a circuit breaker Dj2, a circuit breaker Dj4, generators EG01 and EG02, the respective keys (C6, C7) of which connect them on the bus (03) respectively at the connection points EGA03 and EGB03 interspersed with a bus sectionalizing circuit breaker (Bus-Tie), a transformer (T5) connected to a low voltage load (L6), an aggregate motor (M7).

(2) FIG. 2 is a representation of the proposed process divided in three chronologically ordered phases (phase 1, 2 and 3). For each one of the phases, the ways in which the generators, the VFD converters operate, are presented, in terms of the reactive power exchanged in the microgrid and power factor references; the control algorithm applied: Autonomous (in phase 3) or centralized (implemented in the central controller (CC), in phases 1 and 2); and time frames associated with signaling (Signal between phases 1 and 2) prior to the programmed load activation (direct starting of induction motor), the driven moment (T.sub.phase 2), the period of duration of the electrical transient generated by the load activation (T.sub.phase 2 to T.sub.phase 3) and the return to the permanent regime (return to phase 1 of the process).

(3) FIG. 3 is a graphic representation of the characteristic curve of the modified Volt-VAr control used in the proposed process.

(4) FIG. 4 is a representation of the flowchart of the central controller (CC) algorithm PMS and control embedded in the VFD (modified Volt-Var) in which the calculations of magnitudes involved in the process are explained.

(5) FIG. 5 is a graphic representation of the voltage curve in the main bus as a function of time with indication of the three phases of the process that were described in FIG. 2. The graph has two lines that refer to the base case (microgrid without VFD converters) and the control proposed in the present patent application (centralized combined with modified Volt-VAr), respectively associated with the line colors blue and magenta.

(6) FIG. 6 is a graphic representation of the power factor curve at the terminals of synchronous generators as a function of time with an indication of the three phases of the process that were described in FIG. 2. The graph has two lines that refer to the base case (microgrid without VFD converters) and the control proposed in the present patent application (centralized combined with modified Volt-VAr), respectively associated with the line colors blue and magenta.

(7) FIG. 7 is a graphic representation of the voltage curve in the main bus as a function of time with indication of the three phases of the process that were described in FIG. 2. The graph has two lines that refer to the application of only the central controller (CC) and the control proposed in the present patent application (centralized combined with modified Volt-VAr), respectively associated with the line colors red and magenta.

(8) FIG. 8 is a graphic representation of the power factor curve at the terminals of synchronous generators as a function of time with an indication of the three phases of the process that were described in FIG. 2. The graph has two lines that refer to the application only from the central controller (CC) and the control proposed in the present patent application (centralized combined with modified Volt-VAr), respectively associated with the line colors red and magenta.

(9) FIG. 9 is a graphic representation of the voltage curve in the main bus as a function of time with indication of the three phases of the process that were described in FIG. 2. The graph has two lines that refer to the application of only the classic Volt-VAr control and the control proposed in the present patent application (centralized combined with modified Volt-VAr), respectively associated with line colors black and magenta.

(10) FIG. 10 is a graphic representation of the power factor curve at the terminals of synchronous generators as a function of time with an indication of the three phases of the process that were described in FIG. 2. The graph has two lines that refer to the application only the classic Volt-VAr control and the control proposed in the present patent application (centralized combined with modified Volt-VAr), respectively associated with the line colors black and magenta.

DETAILED TECHNOLOGY DESCRIPTION

(11) A control process for microgrids for voltage regulation on the main bus and power factor (PF) regulation at generator terminals is presented, especially in events scheduled in the microgrid that result in electrical transients, such as direct starting of induction motors (IM). Technology takes advantage of idle capacity of distributed converters (for example: frequency inverters variable frequency drive or VFD) of microgrids making them, in coordinated manner, injecting and/or absorbing reactive power, in addition to exploit the reduced latency of autonomous VFD control during the transient. The Power-Based Control (PBC) technique is used and a modified Volt-VAr function is created applied during the transitional regime. The advantages of the technology are: (1) voltage and power factor regulation, and (2) mitigation of high current demand of the generators during direct starting from IM.

(12) The process for controlling isolated microgrids based on Power-Based Control (PBC) methods and modified Volt-VAr function comprises the following steps: a) identifying the range of power factor (PF) values for the microgrid to be controlled, between the PF*min value and the PF*max value; b) in permanent regime, applying the PBC control method via central controller (CC) and defining the PF references to be used for generators to the maximum value PF*max. VFDs inject reactive power according to calculated quantity (q.sub.ref) via central controller by PBC method; c) identifying the moment in which the scheduled transitional regime will start; d) moments before the start of the scheduled transitional regime identified in step c changing the PF references to be used for generators for their minimum value (PF*min) and programming the VFDs to use the unit as a reference (unitary PF), increasing its availability of reactive power exchange with microgrids within their technically specified limit; e) at the start of the programmed transitional regime, the commands from the central controller are suspended and the VFDs begin to define the reactive power exchanged with the grid autonomously in function of the voltage measured at its connection point through the curve modified Volt-VAr, until the transitional regime ends; f) after the end of the transitional regime, the condition permanent regime described in step b is reestablished: the PBC control method via CC, the PF references to be used for the generators return to the maximum value (PF*max) and the VFD inject reactive power (q.sub.ref) calculated via DC by the PBC method.

(13) The technology can be better understood by the examples below follow, not limiting.

Example 1Application of the Process to an Isolated Microgrid (Computer Simulation)

(14) FIG. 1 shows an isolated microgrid in which the control process being proposed is applied in three phases of operation (as shown in FIG. 2). The results showed below refer to a computer simulation. The phases are defined according to the direct starting moment of induction motors without frequency inverters. Phase 2 is initiated when the event planned start of a large motor is signaled to the central controller. Phase 3 consists of the autonomous operation of the electronic converter via modified Volt-VAr curve. The return to phase 1 takes place after pre-defined machine start times, based on dynamics typical of the elements involved (synchronous generator, induction motor and communication channel between the central control and the frequency inverters). FIG. 2 shows the three phases of the process.

(15) Before describing each of the phases, it is important to explain the invention with its control technique-central controller (CC) with modified Volt-VAr. It is noteworthy that the process has two operation modes, autonomous modified Volt-VAr (phase 3) and centralized control (phases 1 and 2).

(16) The modified Volt-VAr control (FIG. 3) is based on the conventional Volt-VAr. It relates the amount of reactive power exchanged between the frequency inverter and the grid with the voltage measured at the point of system connection (FIG. 1, inverters I1 and I2). When the voltage is within the desired limits (dead band of the modified Volt-VAr curve shown in FIG. 3), the frequency inverter provides a fixed reactive power (q.sub.ref). When the voltage exceeds the limits of the dead band, the exchanged reactive power (capacitive or inductive) increases linearly until reaching a maximum saturation value. This saturation is mandatory in order to guarantee the thermal limits of the frequency inverter. Lastly, it is highlighted that this modified Volt-VAr curve is implemented in the onboard control of each frequency inverter.

(17) The voltage limits of the curve are defined according to the grid codes and electrical system recommendations. For example, the dead band limits are those considered acceptable by the grid operator. Reactive power limits are established in accordance with the thermal capacity of the inverter, based on its nominal values. As represented in FIG. 3, the amount of maximum reactive power, Qmax, is a function of the quadratic subtraction between the nominal power of the frequency inverter and the active power drained to drive the electric motor. In this way, the functionality corresponding to reactive power takes on a secondary priority to the engine drive and operation. This guarantees uninterrupted operation conflicting with the technical and individual limitations of each inverter frequency. The value of q.sub.ref is set through centralized coordinated control, which is based on the PBC technique, and explained below.

(18) Power-based centralized coordinated control (PBC) consists of the coordination of the converters present in the microgrid to achieve a certain power factor at the generator terminals, in permanent regime. FIG. 4(a) illustrates this technique of control on an offshore oil platform, as a study of case. The PMS block is the central controller that measures voltage and current on the main bus (ug and ig) and receives power information (output reactive power, Qi,ufd, and maximum reactive power available, Qi,max) of the i.sup.th frequency inverter distributed in the electrical grid through a communication bus. When measuring and receiving this information, the PMS runs an algorithm (explained in sequence) that returns as output a command coefficient, b. This coefficient is sent to all frequency inverters, that when receiving it, sets the operating q.sub.ref reference of the Volt-VAr curve in function of the previous coefficient b and its maximum available capacity, Qi,max. At the end of this process, the control cycle restarts with sending information packages (Qi,ufd and Qi,max) to the PMS, as represented in the flowchart in FIG. 4(b).

(19) It should be noted that the active power of the converters is determined by the motor itself that the frequency inverter is driving, and the active and reactive power are orthogonal to each other and decoupled. In this way, it is guaranteed that cooperative coordinated control does not restrict or degrade the main operation of the frequency inverter, it only exploits their idle capacity.

(20) The following describes the algorithm implemented in the PMS that returns the command coefficient, b, as output. The variable k represents the kth control cycle of the PMS controller, and the variable i indicates the i.sup.th frequency inverter, which can assume values of 1, 2, . . . , N; where N is the total number of frequency inverters available for the coordinated control. The control algorithm starts with the measurement of voltage and current (ug and ig, as shown in FIG. 4) through the PMS, and the calculation of active and reactive power, calculated by equations (1), (2) and (3).

(21) $\begin{array}{cc}{P}_g=\frac{1}{T}.Math.{}_{t-T}^t\left(v_g.Math.i_g\right)d& \left(1\right)\end{array}$$\begin{array}{cc}{Q}_g=\frac{1}{T}.Math.{}_{t-T}^t\left({\hat{v}}_g.Math.i_g\right)d,& \left(2\right)\end{array}$$\begin{array}{cc}{\hat{v}}_g=.Math.\left(v_{g_f}-\stackrel{\_}{v_{g_f}}\right)=.Math.\left({}_0^t{v}_{g}d-\frac{1}{T}.Math.{}_{t-T}^t{v}_{g_f}d\right)& \left(3\right)\end{array}$

(22) T is the fundamental period of the voltage, and u{circumflex over ()}g is the homo-integral of the measured voltage calculated by equation (3), such that ug is the integral of ug, ug is the average value of ug, and w=2p60 (rad/s) is the angular frequency.

(23) The PMS also receives, through the communication bus, reactive power information, Qi,ufd, and maximum reactive power available, Qi,max of each frequency inverter, and calculates the total reactive power (Qt,ufd) and the total maximum reactive power available (Qt,max) provided by the N frequency inverters during control cycle k, given by equations (4), (5) and (6).

(24) $\begin{array}{cc}{Q}_{t,\mathrm{fvd}}\left(k\right)=\underset{i=1}{\overset{N}{.Math.}}{Q}_{i,\mathrm{vfd}}\left(k\right),& \left(4\right)\end{array}$$\begin{array}{cc}{Q}_{t,\max }\left(k\right)=\underset{i=1}{\overset{N}{.Math.}}{Q}_{i,\max }\left(k\right),& \left(5\right)\end{array}$$\begin{array}{cc}{Q}_{L,t}\left(k\right)=Q_g\left(k\right)+Q_{t,\mathrm{vfd}}\left(k\right),& \left(6\right)\end{array}$

(25) The total reactive power demanded by the microgrid (QL,t), considering a power exchange with synchronous generators is estimated by equation (6). Once the reactive power demand is estimated in the microgrid, the portion of reactive power that the generators must provide, Qg*, for a given power factor reference configured by the offshore electrical system operator, is determined through equations (7), (8) and (9).

(26) $\begin{array}{cc}{Q}_g^{*}\left(k+1\right)=P_g\left(k\right)\mathrm{tg}[a\cos \left({\mathrm{FP}}^{*}\left(k+1\right)\right),& \left(7\right)\end{array}$$\begin{array}{cc}{Q}_t^{*}\left(k+1\right)=Q_{L,t}\left(k\right)-Q_g^{*}\left(k+1\right),& \left(8\right)\end{array}$$\begin{array}{cc}b=\frac{Q_{L,t}\left(k\right)-Q_g^{*}\left(k+1\right)}{Q_{t,\max }\left(k\right)}=\frac{Q_t^{*}\left(k+1\right)}{Q_{t,\max }\left(k\right)},& \left(9\right)\end{array}$

(27) The portion of reactive power that must be supplied by the frequency inverters, Qt*, in the next control cycle (k+1) is given by equation (8). Note that the operation of the next cycle (k+1) is based in estimates in control cycle k. Furthermore, a closed feedback by the communication channel between the central control and the frequency inverters compensate for possible measurement deviations and line losses. Finally, the algorithm ends with the calculation of the command coefficient, b, calculated by equation (9), where b is the reactive power command sent to frequency inverters that will be locally multiplied by Qi,max to define the reference reactive power to be exchanged with the electrical system, q.sub.ref.

(28) $\begin{array}{cc}{q}_{\mathrm{ref}}=b{Q}_{i,\max }.& \left(10\right)\end{array}$

(29) Note that the value of b is limited to [1,1] which guarantees that the thermal limits of the inverters are not exceeded. Furthermore, b can take on positive and negative values, which represent capacitive and inductive reactive exchange with the electrical grid.

(30) Finally, it is highlighted that the stability of centralized coordinated control is guaranteed if the processing time of the algorithm and sending the command coefficient does not exceed the time of a control cycle, which is limited by communication technology used. In other words, sending the coefficient b is limited by the dynamics of the communication system employed. Having clear how the control technique acts, each of the phases of the process, and the two operation modes of coordinated control (autonomous modified Volt-VAr and coordinated control). Details below the three phases of the process represented in FIG. 2.

(31) Phase 1 (operation mode with centralized controlmaximize power factor): Phase 1 corresponds to the permanent regime operation mode called centralized control. In this operation mode, the reference power factor is set to its maximum value (close to unity). This must be defined by a higher management system, tertiary level of control, based in different methods. In this way, during the permanent regime of the system, the load on the generators is minimized and the input active power energy is maximized.

(32) Phase 2 (operation mode with centralized controlPF reduced): This phase of the process marks the transition between the Centralized control operation to autonomous modified Volt-VAr operation mode. This phase begins with signaling to the central controller that a planned event (e.g., departure of a large motor) will happen, through the variable Signal in FIG. 2. In this moment, the power factor reference, PF*max, is changed to a lower value, PF*min, typically between 0.9 and 0.95. This value is defined according to the characteristics of the microgrid. This phase exists to that the action of the converter in reducing voltage sags during engine starting is more effective. This, therefore, reduces the power factor reference of frequency inverters operate with greater availability of Qi,max, allowing better voltage support at the main bus.

(33) Phase 3 (autonomous operating mode with Volt-Var modified): Phase 3 represents the autonomous modified Volt-VAr operation mode used during engine starting. In this operation mode, the frequency inverters no longer follow commands of the central controller and start to define the reactive power exchanged with the grid depending on the voltage measured at its connection point, as FIG. 4. This change is necessary for the converter to act instantly on the main bus without intrinsic latencies of the communication channel. In this way, during engine starting, the converter exchanges reactive power according to the modified Volt-VAr curve. The duration of this phase is pre-defined and varies according to the characteristics of the loads (i.e., motors) of the system. At the end of this phase, the process returns to phase 1, steady state operation, if no other planned events occur subsequently.

Example 2Application of the Process on an Offshore Oil Platform

(34) For the purpose of illustrating the process, an example of application on an offshore oil platform is presented. A microgrid used is shown in FIG. 1 and will be compared with the platform results without any frequency inverter (base case). FIG. 5 shows the voltage profile on the main bus. The invention, in addition to increasing the voltage in regime, also reduces the oscillations at the moment the machine starts when compared to the case without converters.

(35) FIG. 6 shows the power factor at the terminals of the synchronous generators. Note that centralized coordinated control is capable of regulating the power factor in steady state to the pre-defined value (0.95 in this case), that is, the power factor goes from 0.87 to 0.95. This increase in power factor reduces the power reactive capacity of each generator from 15.52 MVAr to 7.93 MVAr, which implies in a reduction in generator load from 0.8834 to 0.7930, representing an energy increase of 10%.

Example 3Individual Comparison Between Conventional PBC and Volt-Var Techniques and the Proposed Technology Via Computational Simulation Results

(36) There are two technologies in the state of the art, which are the Power-Based Control (PBC) and conventional Volt-VAr. Both are explained and compared with the proposed process.

(37) Advantages of the proposed technology compared to the PBC technique: Although the strategy centralized in a central controller (CC) is capable of regulating the power factor at the output terminals of the synchronous generators and provide voltage support, latency and delay intrinsic characteristics of the communication channel delay the action of the converter during scheduled starting of an induction machine. The main advantage of the proposed process in relation to the control strategy proposal essentially consists of the transition from the centralized mode to autonomous mode. In this case, programmed transient voltage regulation becomes more efficient for the proposed process due to the shorter response time of the autonomous converter (power control loops typically operate with a cutoff frequency of tens of Hz) and momentary independence of the communication channel (slow response, in the order of seconds).

(38) In order to visualize these benefits, a simulation comparing the two cases was carried out. FIG. 7 shows the voltage in the main bus and FIG. 8 shows the power factor in the generators. As can be seen, in a permanent regime, the two control techniques present similar results. However, the proposed process is capable of minimizing overvoltages and undervoltages transients in the main bus voltage resulting from starting direct from a three-phase induction motor.

(39) Advantages of the proposed technology in relation to the conventional Volt-Var technique: The proposed process is superior to conventional Volt-VAr technology for electrical systems with large number of inductive loads. The proposed process raises the profile of voltage even under steady state operating conditions with voltage within acceptable limits, which does not happen with conventional Volt-Var. To exemplify these advantages, a simulation comparing the two cases was carried out. FIG. 9 and FIG. 10 present the voltage on the main bus and power factor on the generators where the aforementioned advantages can be visualized. When activating the conventional Volt-VAr still in phase 1, the synchronous generators operate with a power factor of 0.92 in steady state. For the process proposed, the generators operate with configurable power factor (in this case, 0.95) in steady state, which indicates an energy input to the electrical system in relation to the conventional Volt-VAr strategy. In contrast, both strategies contribute significantly to support voltage in a programmed transient regime, since the process proposed also operates in autonomous mode with the Volt-VAr curve.