System and method for paralleling multiple power supplies

Abstract

Systems and methods for managing loads on a power grid are provided. In some embodiments, a load control system includes one or more power sources connected to a power grid. A method includes determining, by a first power generator connected to the power grid, a power average at a first rate, and generating, by the first power generator, a filtered power average. The filtered average includes a power average at a second rate. The filtered power average is used in a control algorithm to balance load sharing of the power sources on the power grid.

Description

[0001] Cross-references to related applications

[0002] This application claims priority and benefit from U.S. Patent Application No. 63 / 032,191, filed May 29, 2020, the entire disclosure of which is hereby incorporated by reference. Technical Field

[0003] This disclosure generally relates to power transmission. More specifically, this disclosure relates to systems and methods for controlling power on a power grid.

[0004] background

[0005] On a power or utility grid, there may be one or more energy sources (e.g., generators, wind turbines, gas turbines, steam turbines) designed to supply power to one or more loads. For example, a grid (e.g., a microgrid, a utility grid, etc.) may include two or more generators connected in parallel with three or more loads. However, microgrids are typically difficult to set up and maintain, requiring specialized skill sets and significant deployment time.

[0006] Overview

[0007] One implementation relates to a method comprising determining an average power value at a first rate by a first generator set connected to a power grid, and generating a filtered average power value by the first generator set. The filtered average power value includes an average power value at a second rate. The filtered average power value is used in a load control algorithm to balance load share among power sources on the power grid. In some embodiments, the control algorithm operates at the second rate. In some embodiments, the second rate is slower than the first rate. In some embodiments, the first generator set may generate the filtered average power value by low-pass filtering the average power value determined at the first rate.

[0008] The method may further include receiving a filtered power average value via a network at a second generator set, and adjusting the power output of the second generator set such that the filtered power average value of the second generator set is aligned with the filtered power average value of the first generator set. Adjusting the power output of the second generator set may include increasing the power output of the second generator set in response to determining that the filtered power average value of the second generator set is lower than that of the first generator set. Adjusting the power output of the second generator set may also include decreasing the power output of the second generator set in response to determining that the filtered power average value of the second generator set is higher than that of the first generator set.

[0009] In some embodiments, the method may further include receiving a filtered average power value via a network at an energy storage system (ESS) and adjusting a set point of the ESS's droop curve at the ESS, wherein the set point affects the ESS's response to electrical parameters sensed on the grid. Adjusting the droop curve set point may include setting the set point to a maximum charging configuration in response to the filtered average power value being below a lower threshold. The lower threshold may indicate that the first generator set is at its minimum load requirement. In some embodiments, adjusting the droop curve set point may include setting the set point to a maximum discharging configuration in response to the filtered average power value being above an upper threshold. Brief description of the attached diagram

[0011] This disclosure will be more fully understood in conjunction with the accompanying drawings, in which similar reference numerals denote similar elements, wherein:

[0012] Figure 1 This is a block diagram illustrating a load control system according to an exemplary embodiment.

[0013] Figure 2 This is a flowchart of a method for controlling power on a power grid according to an exemplary embodiment.

[0014] Figure 3 This is a diagram illustrating the filtering of the average power value of the power supply based on network constraints according to an exemplary embodiment.

[0015] Figure 4A This is a drooping diagram of the ESS according to an exemplary embodiment.

[0016] Figure 4B A sag curve of the ESS according to the first objective is depicted according to an exemplary embodiment.

[0017] Figure 4C A sag curve of the ESS according to the second objective, as shown in an exemplary embodiment, is depicted.

[0018] Figure 5 This is a flowchart of a method for controlling the voltage and hertz (V / Hz) of an ESS according to an exemplary embodiment.

[0019] Figure 6 This is a flowchart of a method for using the ESS power-saving mode according to an exemplary embodiment.

[0020] Figure 7 It is a graph of the droop curve of a one-way power supply according to an exemplary embodiment.

[0021] Detailed description

[0022] Before turning to the accompanying drawings, which illustrate exemplary embodiments in detail, it should be understood that this application is not limited to the details or methods set forth in the specification or shown in the drawings. It should also be understood that terminology is used for descriptive purposes only and should not be considered limiting.

[0023] Referring generally to the accompanying drawings, systems and methods for controlling power on a power grid are provided according to exemplary embodiments. A system includes one or more power sources connected to a power grid. The power grid may also be connected to one or more loads. In some embodiments, the power sources may include generator sets (“gensets”, including combinations of engines and alternators, for example), energy storage systems (ESS), unidirectional power sources (e.g., solar power, wind power), and / or other types of power sources. In some embodiments, the system may include multiple generator sets connected in parallel to the power grid. To ensure that the power output at each generator set is balanced with other generator sets, the generator sets may communicate with each other. However, the bandwidth of some networks may limit the ability of multiple power sources to communicate at a sufficient update rate to maintain the stability of electrical parameters on the power grid and to balance the load shared by each power source in real time. Therefore, to save bandwidth, the load-sharing control algorithm for each power source may be slowed down and attenuated. In this way, each power source can be configured to react via a first algorithm at a first rate (e.g., 50 Hz) to adjust its power output in response to deviations in electrical parameters from nominal values, thereby attempting to maintain grid stability. Each power source can also be configured to react or adjust its power output at a second rate (e.g., 1 Hz) towards a nominally balanced steady state via a load-sharing control algorithm. In this way, transient events are effectively handled in an open-loop manner (e.g., by each power source reacting with minimal influence from the conditions of other power sources), and the load-sharing control algorithm is configured to drive the system towards a nominally balanced steady state at a second rate.

[0024] For example, in one embodiment, the load-sharing control algorithm for the first generator set can be slowed down or attenuated to a second rate (e.g., 1 Hz). The first generator set can be configured to update or recalculate the power average at the first rate (e.g., 50 Hz). The first generator set can use the power average at the first rate and electrical parameters (e.g., frequency and voltage) detected on the grid in a first algorithm to attempt to keep the electrical parameters sensed on the grid at or near nominal values ​​(e.g., 60 Hz and / or 120 Vrms). The first generator set can filter the power average (e.g., updated at 50 Hz) via a low-pass filter having a cutoff frequency half that of the second rate to generate a filtered power average. The filtered power average (e.g., together with other filtered power averages received from other sources and at the second rate) can then be used in the load-sharing control algorithm to adjust the power output of the first generator set, for example, to balance the load sharing of sources within the system. The filtered power average of the first generator set can also be transmitted via the network to other power sources (e.g., which may similarly have load control algorithms operating at or near the second rate), and the elimination of higher frequency components of the power average reduces the amount of bandwidth required on the network. Therefore, the system and method described herein allow multiple power sources to maintain grid stability while also guiding the load towards nominal steady-state equilibrium. In this way, the system and method described herein simplify the integration of power sources into the grid (e.g., microgrids) by reducing the need for high-bandwidth networks and ensuring that all power sources connected to the grid operate optimally according to their respective strengths. Furthermore, the system and method reduce the cost, time, and maintenance required to establish a microgrid while also allowing the integration of renewable power sources (e.g., solar or wind power) into the microgrid.

[0025] refer to Figure 1 A block diagram illustrating a load control system 100 is shown according to an exemplary embodiment. The load control system 100 includes one or more power sources 101a-101c, a power grid 102, and a network 103 (e.g., a communication network). Although three power sources 101a-101c are depicted, it should be understood that in some embodiments, additional or fewer power sources 101a-101c may be implemented in the system 100.

[0026] Power supplies 101a-101c can be connected in parallel to the power grid 102. That is, each of power supplies 101a-101c is configured to be electrically connected to the power grid 102 (e.g., one or more common terminals depending on the power type and specific application). In some embodiments, the respective output terminals of power supplies 101a-101c can be connected to a common bus via a transfer switch, fuse, ATS, or other contacts that can be used to electrically connect and disconnect the respective power supply from the power grid 102. In some embodiments, the power supply may include a generator, generator set, turbine, power plant, solar energy, battery equipment, energy storage system (ESS), or any device or system configured to supply power to the power grid 102.

[0027] Power supplies 101a-101c are communicatively coupled to network 103 and configured to transmit and receive data to and from other power supplies 101a-101c in the system via network 103. In some embodiments, network 103 is a digital communication network. Network 103 may include one or more routers, switches, or other communication hardware configured to route information between power supplies 101a-101c, server systems, and / or the Internet.

[0028] Each of power supplies 101a-101c includes a controller 105. Controller 105 may include one or more processors, memory devices, sensors, and input / output interfaces. The memory may be configured to store machine-executable instructions thereon that, when executed by one or more processors, cause the processors to perform or assist in performing any of the operations, methods, or processes described herein. In some embodiments, the input / output interfaces are configured to allow the respective power supply to communicate with other power supplies via a network or directly. In some embodiments, sensors are configured to monitor various operating parameters of the power supply, such as current output power, frequency or voltage at the output terminals (e.g., electrical parameters for monitoring power on the grid), or other operating parameters of the power supply. Controller 105 is configured to run a load-sharing algorithm. The load-sharing algorithm is typically configured to attempt to maximize the operating parameters of each power supply.

[0029] It should be understood that load-sharing algorithms can include different operating parameters or objectives depending on the specific power source. For example, a generator load-sharing algorithm can be designed to match the current output power of a generator as a percentage of the total generator power capacity (e.g., the average power of the generator) to the current output power of other generators in the system as a percentage of the total generator power capacity (e.g., the average power of other generators). In other words, the generator load-sharing algorithm is configured to adjust the power output of the appropriate generators so that all generators in the system share the load equally according to their respective percentages of output capacity. In this way, the generator load-sharing algorithm reduces the likelihood of some generators overworking and some generators receiving (e.g., not outputting) power, which ensures maximum generator lifespan and efficiency.

[0030] In another example, the energy storage system (ESS) may have different load-sharing algorithm parameters or objectives. For example, in some embodiments, the ESS load-sharing algorithm may dynamically set droop parameters so that the ESS can quickly and efficiently remove power (e.g., charge) when the grid frequency is too high and supply power to the grid when the frequency is too low. The amount of power supplied by the ESS or received from the grid may be expressed as a linear function relative to frequency. In some embodiments, the ESS is configured to dynamically change the linear function (e.g., slope) or setpoint based on the average power of other sources. Reference Figures 4A-4C The droop function of ESS is further discussed.

[0031] In another example, the unidirectional power source (such as wind or solar power) in system 100 may have other algorithmic parameters or objectives. For example, the unidirectional power source may also have a droop parameter; however, an exception is that the unidirectional power source may not receive power from the grid. The unidirectional power source may have a load-sharing control algorithm configured to act according to a droop function. The droop function of the unidirectional power source is typically configured to reduce the output power of the unidirectional power source when the frequency on the grid exceeds a nominal value (e.g., 60 Hz). Reference Figure 7 The droop function of a unidirectional power source was further discussed.

[0032] Therefore, the load-sharing control algorithm for the generator sets is configured to balance the load experienced by each of the generator sets, and the load-sharing control algorithms for the ESS and unidirectional power sources (e.g., renewable power sources) are configured to quickly and effectively absorb transients experienced on grid 102. The load-sharing algorithms also allow for optimal operating parameters for each power source some time after a transient event. For example, after a transient event, each of the generator sets can iteratively or continuously adjust its output power to balance the load, the unidirectional power source can iteratively or continuously operate towards providing the maximum possible power to grid 102 (e.g., the maximum amount of renewable clean power), and the ESS will support the load (if needed), begin recharging, or enter standby mode (e.g., if fully charged and no load support is required). Thus, the system provides a distributed topology of power sources within the grid that can communicate with each other in a low-bandwidth network without requiring a supervisory controller to maintain stability and optimal operating conditions, reducing the cost and complexity of implementing power sources within the grid or microgrid.

[0033] As an example of communication on network 103, in some embodiments, power supplies 101a-101c are configured to determine their respective power averages and transmit their respective power averages to other power supplies 101a-101c within system 100. The power average can be the current power output of the power supply relative to its total power capacity. For example, power supply 101a can be a generator set with a power capacity of 100 kW. The power average of power supply 101a can be the current output (e.g., 1 kW) of the power supply divided by its power capacity (e.g., 10 kW) and expressed as a percentage (e.g., 10%). In some embodiments, a power supply (e.g., a grid-forming generator set) may initially share its output capacity (e.g., kW and / or kVar) with other power supplies, and the power average can be the current power output of the power supply, which can then be used by other power supplies to calculate or determine a corresponding current power output to match the percentage of the power supply's current power output relative to its power output capacity. In some embodiments, the power average can indicate the percentage of load experienced by the corresponding power supply (e.g., a generator set). In some embodiments, power supplies 101a-101c may determine the corresponding power average at a refresh rate of 50 Hz (e.g., the number of times the corresponding power average is updated or recalculated per second). In some embodiments, power supplies 101a-101c may have a refresh rate greater than or less than 50 Hz. Power supplies 101a-101c may filter the power average via a low-pass filter to remove higher frequency components (e.g., transient events) of the power average to create a filtered power average. The filtered power average can be used on the respective power supply for use in a load control algorithm and is also transmitted to other power supplies. For example, in some embodiments, first power supply 101a may update or recalculate the corresponding power average at a rate of 50 Hz (e.g., updating the corresponding power average 50 times per second) and pass the updated or recalculated power average through a low-pass filter (e.g., with a cutoff frequency of 0.5 Hz) to generate a filtered power average, and the filtered power average can be transmitted to other power supplies via a network. In this way, higher frequencies of the power average are filtered out, and the filtered power average can be transmitted via a network. Further discussion on the calculation of average power and filtering will be provided in the reference below. Figure 2 Let's have a discussion.

[0034] In some embodiments, power sources 101a-101c each transmit their respective average power values ​​to the other power sources 101a-101c, and the load control algorithm for each of the power sources 101a-101c can operate to balance the average power values, such that each power source (e.g., a generator set) operates at maximum efficiency. For example, the other generator sets can use the average power values ​​to ensure that each generator set is outputting a similar average power value while also maintaining grid stability (e.g., frequency and voltage stability on grid 102). The local load control algorithm is configured to adjust the power output of the respective generator sets in an attempt to ensure that all generator sets share the load equally (e.g., such that each generator set is outputting a similar percentage of power relative to its respective output capacity). The local load control algorithm can operate cyclically at a second rate (e.g., 1 Hz).

[0035] In some embodiments, the first power source 101a may be designated as a grid former (e.g., a grid-formed generator or a grid-formed ESS). That is, each controller 105 may be configured to identify, communicate, or otherwise designate a particular generator set as a grid-formed generator set. In some embodiments, the largest generator set in the system (e.g., connected to the grid and online) is identified or designated as a grid-formed generator set. In some embodiments, an indication of a grid-formed generator set is sent or programmed into the controller of each generator set. In some embodiments, for example, when multiple generator sets have the same total power capacity, a grid-formed generator may be identified or designated as the generator set with the lowest source address (e.g., the lowest source address among multiple generator sets with the highest total output capacity). In some embodiments, for example, when multiple generator sets have the same total power capacity, a grid-formed generator may be identified or designated as the generator with the highest average power (e.g., the generator subjected to the largest load). In some embodiments, each controller identifies a grid-formed generator set. In some embodiments, a supervisory or master controller may designate a grid-formed generator set and transmit that designation to each of the controllers 105. In some embodiments, such as when no generator set is connected to the grid (e.g., due to anticipated disconnection or because there is no qualified generator set due to generator set being offline), the ESS can be identified or designated as a grid-connector.

[0036] The first power source 101a (e.g., a grid-connected generator) may not run a load control algorithm. More specifically, the first power source 101a can maintain the frequency and voltage of the power on grid 102 within a predetermined range (e.g., + / - 10% for 60 Hz, + / - 10% for 120 volts). The first power source 101a (e.g., a grid-connector) can then determine an average power value (e.g., the average power value of the first power source 101a) and transmit this average power value via network 103 to the other power sources 101b-101c in the system. The other power sources can then use the received average power value from the first power source 101a to balance load sharing and maintain grid stability. In some embodiments, the grid-connected generator can locally limit the current power output at a high execution rate (e.g., 50 Hz) to prevent reverse power while balancing and to control ramp load and ramp unload. In some embodiments, each of the power supplies 101a-101c can locally limit the current power output at a high execution rate (e.g., 50 Hz) to prevent reverse power while balancing and to control ramp loading and ramp unloading.

[0037] Now for reference Figure 2 The diagram illustrates a flowchart of a method 200 for controlling power on a power grid according to an exemplary embodiment. Specifically, method 200 involves achieving a nominally balanced steady-state operating point (e.g., stable and optimized operating parameters for each power source) in the power grid by communicating over a network (e.g., a network with limited bandwidth due to the state of current technology and the cost of implementing state-of-the-art networks).

[0038] In operation 201, a first power source determines a power average (e.g., the power average of the first power source). The first power source is connected to a power grid (e.g., a microgrid) and may be experiencing a load. The controller of the first power source may determine the current output power of the first power source and determine the power average. The power average may be a percentage of the current output power of the first power source and the total capacity of the first power source. Furthermore, the controller of the first power source may filter the power average, such as by slowing down the refresh rate of the power average. In some embodiments, the controller of the first power source filters the power average (e.g., and / or slows down the refresh calculation of the power average) to a refresh rate lower than the desired network rate. For example, in some embodiments, the network communicatively coupled to the first power source may have limited bandwidth unsuitable for transmitting the power average at a high frequency (e.g., 50 Hz). The first power source may slow down the refresh or recalculation rate of the power average to a rate suitable for transmission on the network (e.g., 1 Hz). Furthermore, in some embodiments, the controller may filter the power average via a low-pass filter having a cutoff frequency equal to two times the network rate (e.g., the rate at which the network can transmit without bandwidth issues). For example, if data from a power source (e.g., average power) can be transmitted and received over a network at 1 Hz, the first power source can filter the average power with a low-pass filter having a cutoff frequency of 0.5 Hz before transmitting it. (See reference...) Figure 3 The filtering of the power average value is discussed in further detail.

[0039] In operation 202, a power average (e.g., a determined or filtered power average) is transmitted via a network from a first power source to other power sources within the system. This transmission may be digital via the network. In some embodiments, the power average is received by each of the other power sources and used to control power on the grid (e.g., load sharing). In some embodiments, other power sources may similarly determine their own power averages (e.g., filtered power averages to save bandwidth) and may transmit these power averages to multiple power sources on the grid (e.g., including the first power source). In some embodiments, the first power source may also include an indication of its identity, enabling other power sources to identify which power source within the system the power average belongs to.

[0040] In operation 203, the first power source and other power sources use power averages to achieve optimal steady-state operating conditions (e.g., via a local load control algorithm operating at or near a second rate). That is, the power sources can use power averages in the local load control algorithm to attempt and match power averages, such that each power source shares the load of the grid. For example, other generators in the system can operate to achieve a nominally balanced steady-state operating point based on filtered power averages received from one or more other power sources and the local load control algorithm. In some embodiments, the first power source can be a grid-connected generator. Grid-connected generators can regulate the frequency and voltage on the grid without adjusting the load sharing of the grid-connected generators.

[0041] For example, other generator sets within the system may receive an average power value (e.g., a filtered average power value) from a first power source and use the filtered average power value in a local load control algorithm (e.g., operating at or near a second rate) to attempt to match the filtered average power value of the first power source (e.g., a grid-connected generator set). In an example where one or more of the other generator sets have an average power value higher than the average power value of the first power source, the one or more other generator sets may reduce their respective power outputs to attempt to match the average power value of the first power source. In response, the first power source may detect or determine that the frequency and voltage on the grid are decreasing and then begin to output more power to balance load sharing (e.g., by increasing the average power value of the first power source and decreasing the average power value of one or more of the other generator sets). In another example where one or more of the other generator sets have an average power value lower than the average power value of the first power source, the one or more other generator sets may increase their respective power outputs. In response, the first power source can detect or determine that the frequency and voltage on the grid are rising, and then begin to reduce the power output of the first power source to balance load sharing (e.g., by increasing the average power of other generator sets and decreasing the average power of the first power source). In some embodiments, other generator sets (e.g., those that are not grid-connected generator sets) can similarly calculate and filter their respective average power values ​​before comparing them with the average power value of the first power source (e.g., the average power value of grid-connected generator sets).

[0042] Now for reference Figure 3 Figure 300 illustrates the determination of the average power value of a power supply based on network constraints according to an exemplary embodiment. In other words, the figure depicts the processing flow of the power supply controller when determining an average power value suitable for transmission over a limited bandwidth network.

[0043] The power supply can calculate a power average 301 at a first rate (e.g., 50 Hz), which includes frequencies too high for the network bandwidth to accommodate. The network rate is the rate at which the power average can be transmitted over a network with limited bandwidth. Therefore, the load control algorithm can be configured to accommodate a second rate (e.g., selected due to network bandwidth constraints) that is compatible with the network rate, and the filtered power average can be filtered via a low-pass filter having a cutoff frequency half that of the second rate (e.g., which can be selected as the network rate). The power supply feeds the power average 301 at the first rate into a low-pass filter 302 to remove frequency components greater than the network rate (e.g., 1 Hz) divided by 2. The low-pass filter 302 can be a first-order low-pass filter with a cutoff frequency of the network rate divided by 2. The low-pass filter 302 can then output the power average at the second rate (e.g., the filtered power average) via network 303 to other power supplies and to a load-sharing control algorithm 304 on the power supply. Filtering ensures that frequency components above the Nyquist frequency are removed before transmission, which improves the accuracy of the transmitted power average.

[0044] Now refer to Figure 4A The diagram illustrates a droop profile 400 of an ESS according to an exemplary embodiment. Specifically, droop profile 400 depicts various droop curve setpoints of the ESS, which can be dynamically set or adjusted based on the average power received from the power source (e.g., a grid-connected generator set). Dynamic ESS droop curve setpoints allow the ESS to support the generator set (e.g., absorb load by charging when the generator set is at or near minimum load requirements, or provide power when the generator set is at or near maximum load capacity) and also absorb transient events experienced on the grid.

[0045] The droop graph 400 has a y-axis 401 indicating the output power of the ESS and an x-axis 402 indicating electrical parameters (e.g., frequency and voltage) of the sensed power on the grid. The droop graph 400 includes an operating range 403. The operating range 403 indicates the potential power output response of the ESS relative to the electrical parameters of the grid. For example, for illustrative purposes, a first curve 410, a second curve 411, and a third curve 412 within the operating range 403 are discussed in detail. Curves 410, 411, and 412 indicate how the ESS will respond (e.g., output power or receive power from the grid) to a detected deviation of the grid's frequency and / or voltage from its nominal value. It should be understood that curves 410, 411, and 412 are shown only by way of example, and a continuum of potential curves exists within the operating range. For additional reference, the top 480 of the operating range 403 indicates the maximum power output from the ESS to the grid (e.g., 100%), and the bottom 481 of the operating range 403 indicates the maximum power consumption of the ESS from the grid (e.g., -100%).

[0046] The first curve 410 indicates the maximum charge droop state, where the ESS is configured to charge (e.g., receive power from the grid) even if the grid frequency and / or voltage are below nominal values. However, if the grid frequency and voltage fall below a lower threshold 490, the ESS will output power. The lower threshold can be -10% of the nominal frequency and voltage values.

[0047] The second curve 411 indicates a zero-power droop state, where the ESS is configured to charge (e.g., receive power from the grid) if the grid frequency and / or voltage are above nominal values, and to supply power to the grid if the grid frequency and / or voltage are below nominal values. The nominal frequency and / or voltage can be set based on the specific application or location of the grid.

[0048] The third curve 412 indicates the maximum discharge state, where the ESS is configured to discharge (e.g., output power to the grid) even if the grid frequency and / or voltage is above a nominal value. However, if the grid frequency and / or voltage is above an upper threshold 491, the ESS will charge (e.g., receive power from the grid). The lower threshold can be +10% of the nominal frequency and / or voltage value.

[0049] The ESS can dynamically select which curve 410, 411, or 412 it will follow based on the average power received from the power source (e.g., a grid-connected generator set). For example, if the grid-connected generator set has a low average power (e.g., indicating that the generator set is not currently outputting a significant amount of power, which may be close to violating the minimum load threshold of the grid-connected generator set), the ESS can dynamically or automatically choose to follow the first curve 410, allowing the ESS to charge and provide additional load. In another example, if the grid-connected generator set has a median average power (e.g., indicating that the generator set is outputting a median amount of power relative to its capacity), the ESS can dynamically or automatically choose to follow the second curve 411, such that when the frequency and voltage are at nominal values ​​but still available to support transient events, the ESS will not receive or output power. In another example, if the grid-connected generator set has a high average power (e.g., indicating that the generator set is outputting a high amount of power relative to its capacity), the ESS can dynamically or automatically choose to follow the third curve 412 such that the ESS is outputting the maximum amount of power when the frequency and voltage are at nominal values ​​in an attempt to support the grid-connected generator set (e.g., and thereby support other generator sets, as they are also iteratively working toward the nominal operating setpoint or equilibrium state).

[0050] The slopes of the first curve 410, the second curve 411, and the third curve 412 can be set or dynamically changed to be sufficiently steep to achieve grid stability. The steepness of the curves can depend on the specific grid being implemented. For example, for applications where the ESS is deployed in a system with anticipated large transient events, the steepness of the curves can be increased to ensure that the ESS responds appropriately to large transient events. In other embodiments, the curves can be non-linear. In some embodiments, the ESS can receive commands from a supervisory controller or other device to set the curves that the ESS will follow. In some embodiments, the ESS can limit instantaneous power output based on the system's short-term failure curve. For example, the ESS can limit instantaneous power by changing the slope of the drooping curve to exceed the system's rating (e.g., changing the slope to zero or near zero, exceeding the system's rating).

[0051] For illustrative purposes, please refer to the following: Figure 4B and Figure 4C . Figure 4B A sag curve 450 of the ESS according to the first objective is depicted according to an exemplary embodiment. Figure 4CA droop curve 460 of the ESS according to a second target is depicted according to an exemplary embodiment. In some embodiments, the target of the ESS can be set via user input or command. For example, user input or command can be transmitted to the ESS via a network through a device (e.g., a supervisory controller) or input via a user interface on the ESS controller. The target can include multiple different settings for kilowatt (kW) output target and / or kilovolt-ampere (kVar) reactance output target.

[0052] For example, the kW output target for an ESS can include charging based on generator constraint settings, maximum charging settings, follow-demand settings, maximum discharging settings, and / or an output kW setpoint. Charging based on generator constraint settings can instruct the ESS to charge the ESS while maintaining a target that limits the generator power output determined by an upper limit threshold (e.g., a predefined generator kW upper limit constraint setpoint). For example, in charging based on generator constraint settings, the ESS can determine or access a droop curve that allows the ESS to charge (e.g., by consuming the difference) when the average power of the generator set (e.g., the grid-connector) is below the upper limit threshold. Maximum charging settings can instruct the ESS to charge as quickly as possible without overloading the generator set. For example, in maximum charging settings, the ESS can determine a droop curve (e.g., droop curve 410) that specifies that the ESS uses all available kW in the generator set for charging without overloading it. The average power of the generator set can be used to determine the amount of kW available for charging. In the maximum discharge setting, the ESS can determine or access a droop curve (e.g., curve 412) that allows the ESS to output as much power as possible up to zero (e.g., or a predefined lower generator load limit) generator load (e.g., indicated to the ESS by the power average). In the output kW setpoint setting, the ESS can maintain the desired or predefined kW output.

[0053] As an illustrative example, Figure 4BA droop curve 450 is depicted in a demand-following setting. Curve 450 includes a y-axis 455 indicating the kW output of the ESS in the demand-following setting, an x-axis 456 indicating the system load kW, and a droop curve 459. The system load kW can be indicated to the ESS via the average power received from the generator set (e.g., grid-connector), or determined by sensing at the output. In the demand-following setting, the ESS can determine or access the droop curve such that the ESS limits the generator set power (e.g., determined by the ESS based on the average power) within a lower threshold 451 (e.g., a predefined lower constraint setpoint for generator kW) and an upper threshold 452. If the average power exceeds the upper threshold 452 or falls below the lower threshold 451, the ESS will attempt to supply or absorb the difference to maintain the average power of the generator set within the lower and upper thresholds. The shape of the droop curve (e.g., the droop curve on curve 450) can be flattened at + / -100% to limit the output to a safe level regardless of bus fluctuations.

[0054] Furthermore, the target for the kVar output of the ESS can include a power factor correction setting and / or an output kVar setpoint. At the output kVar setpoint, the ESS can maintain or attempt to maintain a desired predefined kVar output. Additionally, as an illustrative example, Figure 4C A droop curve 460 is depicted in the demand-following setting. Curve 460 includes a y-axis 465 indicating the kVar output of the ESS in the power factor correction setting, an x-axis 466 indicating the system load kVar, and a droop curve 469. In the power factor correction setting, the ESS can supply kVar to the load to maintain or attempt to maintain the generator set's unity power factor. Furthermore, the ESS can limit its kVar output so as not to exceed the ESS's kVar rating (e.g., indicated by a flat droop curve at the maximum and minimum values). It should be understood that when the ESS is constrained by the upper limit of its hardware rating, kW demand can take precedence over kVar demand or target. At the output kVar setpoint, the ESS can maintain or attempt to maintain a desired predefined kVar output.

[0055] Now for reference Figure 5A flowchart of a method 500 for voltage and hertz (V / Hz) control of an ESS according to an exemplary embodiment is shown. Method 500 describes a method for controlling the output electrical parameters (e.g., voltage and frequency) of the ESS in response to a transient event. Specifically, a solid-state inverter (e.g., as an inverter on the ESS) outputs frequency and voltage according to set parameters (e.g., the nominal value of the grid) within a controller. However, when a transient event occurs on the grid (e.g., the on-state of a large load), the generators on the grid experience increased load and decelerate (e.g., thereby reducing the output frequency and / or voltage of power on the grid). As a result, the ESS reacts and outputs power at the nominal value of the grid, which may cause the ESS to trip, fail, or otherwise disconnect if the load is too large (e.g., because the ESS attempts to absorb the entire load). Disconnection of the ESS during a transient event may cause other ESSs on the grid to also fail, trip, or disconnect, resulting in a cascading failure. Therefore, Method 500 describes a method for dynamically and automatically reducing the output voltage and / or frequency of the ESS to simulate the generator set's response to transient events in order to prevent tripping during transient events. Furthermore, the coordinated V / Hz response between the ESS and the generator set can also reduce the amount of reactive power flow between power sources and improve the overall system response.

[0056] In operation 501, the ESS accesses the voltage and frequency responses of generator sets on the grid. For example, the ESS may receive information from grid-connected generator sets regarding power capacity, average power, or a function of how a transient event might affect the grid-connected generator sets (e.g., or the entire system). In some embodiments, the voltage and frequency responses of generator sets on the grid may be stored in memory on the ESS. In some embodiments, the ESS may dynamically calculate or estimate, based on power capacity, average power, or other information about the generator sets, a function of how a transient event might affect the voltage and / or frequency output of generator sets on the grid.

[0057] In operation 502, the ESS can detect that a transient event is occurring on the power grid. For example, the ESS can detect that the frequency and voltage are deviating positively from their nominal values. In some embodiments, the ESS can determine the extent of the transient event (e.g., the limits of the transient event) based on the amount by which the frequency and voltage deviate from their nominal values. In some embodiments, the ESS can receive, via communication from another device such as a supervisory controller, an indication that a large transient event is occurring or about to occur (e.g., an indication that a large load is about to be connected to the power grid). In response, for example, to prepare for a large load connection (e.g., a load of a predefined size), the ESS can load the generator set as much as possible or by a predefined amount based on the size of the load (e.g., by outputting less power or consuming power and thus increasing the power output of the generator set to maintain grid stability) to avoid hysteresis in the generator set's horsepower response. Once the large load is connected, the ESS can (e.g., by detecting conditions on the power grid) detect that the large load has been connected or receive an indication from a device (e.g., a supervisory controller) and begin to increase the amount of power output from the ESS to absorb the transient event caused by the large load connection. For example, an ESS can automatically switch from charging to discharging, or automatically increase the amount of power discharge to allow for the connection of a large load, so that the generator load (e.g., the increased generator load) remains relatively constant. For example, the exact amount of power output change can be based on a predefined range of the ESS's power output capacity and the size of the large load (e.g., a received indication). In this way, large transient events (e.g., the addition of a large load) can be proactively handled by the power sources on the grid, which improves grid stability.

[0058] In operation 503, the ESS can adjust the frequency and voltage of its output power. For example, the ESS can adjust the frequency and voltage of its output power to simulate the frequency and voltage response of a generator set. In this way, the ESS can provide necessary support to the grid during transient events while reducing reactive power flow between power sources and also reducing the likelihood of faults that could cause the ESS to disconnect. The ESS can adjust the frequency and voltage of its output power according to a function estimated, determined, or accessed in operation 501. In some embodiments, the ESS may not follow a function; instead, the ESS can match the detected frequency and voltage on the grid and output maximum power while continuously (e.g., iteratively) adjusting the ESS's output frequency and voltage to match the grid's current frequency and voltage output (e.g., the grid's current frequency and voltage output should return to nominal when the generator set has time to react and catch up). In some embodiments, the frequency and voltage of the ESS's output may have a minimum threshold (e.g., -10%) that the ESS will not fall below (e.g., attempt to match below) regardless of the detected frequency and voltage on the grid.

[0059] Now for reference Figure 6 A flowchart of a method 600 for an ESS power-saving mode according to an exemplary embodiment is shown. That is, in some embodiments, the ESS may include a power-saving mode. The power-saving mode may be a mode during operation (e.g., when connected to the power grid), wherein the ESS is available but gated so that it neither outputs nor receives power. In other words, in power-saving mode, the ESS may be connected to the power grid (e.g., via a bus) and monitor electrical parameters on the grid, but is in a dormant state to avoid switching losses.

[0060] In operation 601, the ESS determines that a power-saving mode should be activated. The ESS can automatically enter a power-saving mode when it detects that no power is needed from the ESS and no power is required to charge the ESS within a predefined time period. In some embodiments, the ESS can determine that a power-saving mode should be activated in response to receiving manual input or a command from a user or the user's computing device. In some embodiments, the ESS can determine that other power sources are connected to the grid, and that the power demand on the ESS is below a lower threshold (e.g., a predefined value) or within a predefined low range (e.g., 0-20% of capacity power output), and in response, the ESS can determine that it is able to stop outputting power and enter a power-saving mode. For example, the ESS can determine that other power sources are connected to the grid by receiving an average power value from other power sources (e.g., a filtered average power value from a grid-connected generator set).

[0061] In operation 602, in response to the determination made in operation 601, the ESS enters a power-saving mode. The power-saving mode can gate the ESS from the power grid, which can reduce parasitic power consumption. In some embodiments, the ESS can disconnect a switch, electrically disconnecting the ESS from the power grid. However, while in power-saving mode, the ESS can monitor the frequency and voltage of the power on the grid.

[0062] In operation 603, the ESS can determine that it is needed to achieve grid stability and reconnect to the grid. For example, the ESS can monitor whether the frequency and voltage of the power on the grid deviate from the nominal range (e.g., deviate from the nominal value by + / - 3%). In response to detecting that the frequency and / or voltage have deviated from the nominal range, the ESS can be gated back to the grid until grid stability is restored and the ESS power demand has subsided (e.g., the grid frequency and voltage return to the nominal range). For example, in operation 603, the ESS can cause a switch to close, which reconnects the ESS to the grid, allowing the ESS to provide grid stability by outputting or receiving power.

[0063] Now for reference Figure 7The graph 700 depicts a unidirectional power source. For example, a unidirectional power source can include solar power, wind power, or other renewable power sources configured to connect to and output power to the grid. In other words, a unidirectional power source can be configured only to output power to the grid without receiving power from it.

[0064] The graph 700 includes a y-axis 701 indicating the power output level of the unidirectional power source (e.g., in %kW and %kVar) and an x-axis 702 indicating the power value on the grid (e.g., the frequency and / or voltage of the power). The graph 700 also includes a droop curve 703 indicating the response of the unidirectional power source's power output level relative to a detected, monitored, or determined power value on the grid. That is, the droop curve 703 can be set or dynamically determined within the controller of the unidirectional power source, such that the unidirectional power source is configured to maximize the power output of the power source as long as power is being consumed by an ESS or load on the grid.

[0065] For example, when a unidirectional power source determines or detects that the frequency and / or voltage on the grid is at its nominal value (e.g., 60 Hz and / or 240 V), the unidirectional power source is configured to output the maximum possible amount of power (e.g., for temporary overload, which can support grid stability) (e.g., 100%) to the grid. In this way, when the grid is operating stably (e.g., operating at nominal power), the unidirectional power source (e.g., a renewable power source) outputs the maximum amount of renewable power to the grid, ensuring that the grid is supported by as much renewable energy as possible. The droop curve 703 indicates that when the frequency or voltage exceeds its nominal value, the unidirectional power source will reduce its output power, allowing it to respond to grid transients and thus maintain grid stability. It should be understood that the slope of the droop curve 703 or the exact function of the droop curve 703 can be set or dynamically adjusted to ensure grid stability is maintained. For example, the slope of the droop curve exceeding the nominal value can be increased in certain implementations (e.g., implementations with large expected transient phenomena), or the slope of the droop curve exceeding the nominal value can be decreased in other implementations. Following a transient event, a unidirectional power source can iteratively or continuously adjust its power output back to its maximum value (e.g., 100%) to ensure that the maximum amount of renewable power is being used, which increases grid efficiency. In this way, a unidirectional power source can actively push power onto (e.g., output more power to) the grid, allowing other power sources to reduce their corresponding output power and maximize grid efficiency. For example, power sources (e.g., unidirectional power sources and / or generator sets) can implement locally constrained reverse power flow control algorithms to adapt to slower networks with higher latency (e.g., due to limited bandwidth) and actively operate towards a nominal equilibrium steady state without communicating over the network.

[0066] Similar to the user-defined or supervisory-defined targets explained above with reference to ESS, a unidirectional power supply can also have targets input by the user or defined via a supervisory controller. For example, a unidirectional power supply can have a kW target for maximum export and / or output kW setpoint. Furthermore, a unidirectional power supply can have a kVar target for power factor correction or output kVar setpoint settings. A unidirectional power supply with a maximum export kW setting can target 100% power output. However, as described below, if the generator set is below a lower load limit or its power output is 0%, it may not output 100% power output. A unidirectional power supply with an output kW setting can output a predefined kW amplitude. Similarly, a unidirectional power supply with an output kVar setting can output or attempt to output a predefined kVar amplitude. Furthermore, a unidirectional power supply with a power factor correction setting can output as much kVar as possible in an attempt to maintain unity power factor of the generator set.

[0067] As another example, a unidirectional power supply can determine the average power value, which can then be filtered to a suitable network rate (e.g., as referenced). Figure 2 and Figure 3 (As explained), and transmits the power average to other power sources. Other power sources can use the power average of the unidirectional power source to reduce their power output (e.g., if the unidirectional power average indicates that the unidirectional power source is not outputting more than 50% of the power average), so that the unidirectional power source will respond to the reduced power by increasing its power output to the maximum extent (e.g., as a drop in frequency and / or voltage on the grid is sensed or determined by the unidirectional power source), thereby maximizing the grid efficiency again.

[0068] As another example, a unidirectional power source can receive power averages (e.g., filtered power averages) from one or more other power sources (e.g., from a grid-connected generator set) and use these power averages to maximize operating conditions. For example, the power averages might indicate that the grid-connected generator set is at or near a lower threshold indicating the generator set's minimum load requirement, or that the generator set is not outputting any power (e.g., the power average is 0). In response, the unidirectional power source (e.g., an inverter for a solar or wind power source) can reduce the power output of the solar or wind power source, such as to prevent reverse power flow to the generator set. For example, the unidirectional power source can receive power averages, determine that the power average is below a threshold (e.g., a predefined threshold), and, in response, reduce the unidirectional power source's power output so that the generator set does not receive reverse power flow or is above the minimum load threshold. The unidirectional power source can iteratively reduce its power output in response to each received power average, or it can reduce the power output by an amount related to the amount by which the power average is below the threshold. In this way, the unidirectional power source can use the power averages to maximize the operating conditions and lifetime of the generator set or other power sources.

[0069] The present disclosure has been described above with reference to the accompanying drawings. These drawings illustrate certain details of specific embodiments of systems, methods, and procedures for implementing the present disclosure. However, the description of the present disclosure with reference to the drawings should not be construed as imposing any limitations that may exist in the drawings. The present disclosure contemplates methods, systems, and program products on any machine-readable medium used to perform its operations. Embodiments of the present disclosure may be implemented using existing computer processors, or by a dedicated computer processor incorporated for this or another purpose, or by a hardwired system. No element of the claims herein should be interpreted pursuant to paragraph 6 of 35 U.S.SC §112 unless the element is expressly enumerated using the phrase “means for…”. Furthermore, no element, component, or method step in the present disclosure is intended to be exclusive to the public, regardless of whether the element, component, or method step is expressly enumerated in the claims.

[0070] As described above, embodiments within the scope of this disclosure include program products comprising a machine-readable storage medium for carrying or storing machine-executable instructions or data structures thereon. Such a machine-readable storage medium can be any available medium accessible by a computer or other machine having a processor. As examples, such a machine-readable storage medium may include RAM, ROM, EPROM, EEPROM, CD ROM or other optical disc storage devices, magnetic disk storage devices or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of machine-executable instructions or data structures and is accessible by a computer or other machine having a processor. Combinations of the foregoing are also included within the scope of machine-readable storage media. Machine-executable instructions include, for example, instructions and data that cause a computing device or machine to perform a function or set of functions. As referred to herein, machine- or computer-readable storage media do not include transient media (i.e., signals in space).

[0071] Embodiments of this disclosure are described in the general context of method steps, which may be implemented in one embodiment by a program product including, for example, machine-executable instructions, such as program code, in the form of program modules executed by a machine in a networked environment. Typically, program modules include routines, programs, objects, components, data structures, etc., that perform a particular task or implement a particular abstract data type. The machine-executable instructions, associated data structures, and program modules represent examples of program code for performing steps of the methods disclosed herein. A particular order of such executable instructions or associated data structures represents examples of corresponding actions for implementing the functionality described in these steps.

[0072] Embodiments of this disclosure can be implemented in a networked environment using a logical connection to one or more remote computers having processors. Logical connections can include local area networks (LANs) and wide area networks (WANs), presented herein as examples and not limitations. Such networked environments are common in office-wide or enterprise-wide computer networks, intranets, and the Internet, and can use a variety of different communication protocols. Those skilled in the art will recognize that such networked computing environments typically include many types of computer system configurations, including personal computers, handheld devices, multiprocessor systems, microprocessor-based or programmable consumer electronics devices, network PCs, servers, minicomputers, mainframes, etc. Embodiments of this disclosure can also be implemented in a distributed computing environment, where tasks are performed by local and remote processing devices linked via a communication network (via a hardwired link, a wireless link, or a combination of hardwired and wireless links). In a distributed computing environment, program modules can reside in local and remote memory storage devices.

[0073] An exemplary system for implementing an entire system or part thereof may include a computing device including, for example, a processing unit, system memory, and a system bus that couples various system components, including the system memory, to the processing unit. The system memory may include read-only memory (ROM) and random access memory (RAM) or other non-transitory storage media. The computer may also include a magnetic hard disk drive for reading from and writing to a magnetic hard disk, a disk drive for reading from or writing to a removable disk, and an optical disk drive for reading from or writing to a removable optical disc, such as a CD-ROM or other optical media. The drive and its associated machine-readable medium provide the computer with non-volatile storage of machine-executable instructions, data structures, program modules, and other data.

[0074] It should be noted that while the flowcharts provided herein illustrate a specific order of method steps, it should be understood that the order of these steps may differ from that depicted. Two or more steps may also be performed simultaneously or partially simultaneously. This variation will depend on the chosen software and hardware system and the designer's choices. It should be understood that all such variations are within the scope of this disclosure. Similarly, the software and web implementations of this disclosure can be implemented using standard programming techniques with rule-based logic and other logic to perform various database search steps, association steps, comparison steps, and decision steps. It should also be noted that the term "component," as used herein and in the claims, is intended to include implementations using one or more lines of software code and / or hardware implementations and / or devices for receiving manual input.

[0075] The foregoing description of embodiments of the present disclosure is presented for illustrative and descriptive purposes. It is not intended to be exhaustive or to limit the disclosure to the precise forms disclosed, and modifications and variations are possible, or may be derived from practice, in accordance with the foregoing teachings. Embodiments have been chosen and described to explain the principles of the disclosure and its practical application, enabling those skilled in the art to utilize the disclosure in various embodiments and with various modifications suited to the particular purpose contemplated.

Claims

1. A method for controlling power on a power grid, comprising: The power average value at a first rate is determined by a first generator set connected to the power grid, wherein the power average value includes the current power output of the first generator set relative to the total power capacity of the first generator set; The first generator set generates a filtered power average by low-pass filtering the power average at the first rate, wherein the filtered power average includes the power average at a second rate, and wherein the filtered power average is used in a load control algorithm to balance the load sharing of power sources on the power grid, wherein the second rate is slower than the first rate; and The first generator set transmits the filtered power average at the second rate to one or more other power sources connected to the power grid via a network.

2. The method according to claim 1, further comprising: The filtered average power is received via the network at the second generator set; as well as Adjust the power output of the second generator set at the second generator set so that the filtered average power of the second generator set is aligned with the filtered average power of the first generator set.

3. The method according to claim 2, wherein, Adjusting the power output of the second generator set includes increasing the power output of the second generator set in response to determining that the filtered power average of the second generator set is lower than the filtered power average of the first generator set.

4. The method according to claim 2, wherein, Adjusting the power output of the second generator set includes reducing the power output of the second generator set in response to determining that the power average percentage of the second generator set is higher than the power average percentage of the first generator set.

5. The method according to claim 1, further comprising: The filtered average power is received via the network at the energy storage system (ESS); as well as The set point of the sag curve of the ESS is adjusted at the ESS, wherein the set point affects the response of the ESS to electrical parameters sensed on the power grid.

6. The method according to claim 5, wherein, Adjusting the setpoint of the droop curve includes setting the setpoint to the maximum charging configuration in response to the filtered average power value being below a lower threshold.

7. The method according to claim 6, wherein, The lower limit threshold indicates that the first generator set is at its minimum load requirement.

8. The method according to claim 5, wherein, Adjusting the setpoint of the droop curve includes setting the setpoint to the maximum discharge configuration in response to the filtered average power value being higher than an upper limit threshold.

9. The method according to claim 1, wherein, Generating the filtered power average includes low-pass filtering the power average determined at the first rate based on a cutoff frequency associated with the transmission rate of the network.

10. A power source configured to be coupled to a power grid, the power source comprising: The controller is configured to: Determine the average power value of the power supply at a first rate, wherein the average power value includes the current power output of the power supply relative to the total power capacity of the power supply; A filtered power average is generated by low-pass filtering the power average at the first rate, wherein the filtered power average includes the power average at the second rate, and wherein the filtered power average is used in a load control algorithm to balance the load distribution of power sources on the power grid, wherein the second rate is slower than the first rate; and The average power at the second rate is transmitted via a network to one or more other power sources connected to the power grid.

11. The power supply according to claim 10, wherein, The power average value at the second rate is configured to cause the second generator set to adjust the power output of the second generator set such that the filtered power average value of the second generator set is adjusted toward the filtered power average value of the power source.

12. The power supply according to claim 10, wherein, Generating the filtered power average includes low-pass filtering the power average determined at the first rate based on a cutoff frequency associated with the transmission rate of the network.

13. A system for controlling power on a power grid, comprising: A first power source, configured to be coupled to the power grid, the first power source comprising: A first controller, configured to: Determine the average power value of the first power supply at a first rate, wherein the average power value includes the current power output of the first power supply relative to the total power capacity of the first power supply; A filtered power average is generated by low-pass filtering the power average at the first rate, wherein the filtered power average includes the power average at the second rate, and wherein the filtered power average is used in a load control algorithm to balance the load distribution of power sources on the power grid, wherein the second rate is slower than the first rate; and The average power at the second rate is transmitted via a network to one or more other power sources connected to the power grid.

14. The system of claim 13, further comprising a second power source configured to be coupled to the power grid and communicatively coupled to the first power source, the second power source comprising: The second controller is configured to: Receive the filtered average power value of the first power source; as well as Adjust the power output of the second power supply so that the filtered average power of the second power supply is aligned with the filtered average power of the first power supply.

15. The system according to claim 14, wherein, Adjusting the power output of the second power supply includes increasing the power output of the second power supply in response to determining that the filtered power average of the second power supply is lower than the filtered power average of the first power supply.

16. The system of claim 13, further comprising an energy storage system (ESS) coupled to the power grid and communicatively coupled to the first power source, the ESS comprising: The second controller is configured to: The filtered power average is received via the network; as well as Adjust the setpoint of the droop curve of the ESS, wherein the setpoint affects the response of the ESS to electrical parameters sensed on the power grid.

17. The system according to claim 16, wherein, Adjusting the setpoint of the droop curve includes setting the setpoint to the maximum charging configuration via the second controller in response to the filtered average power value falling below a lower threshold.

Images