A wind turbine with a reflective memory system

ES 2 633 166 T3

DESCRIPTION

A wind turbine with a reflective memory system

Technical field

The present invention relates generally to a wind turbine.

Background

In recent years, the demand for green energy has increased. Wind turbines, which convert energy from the wind into electrical energy, are environmentally friendly energy media. As such, confidence in wind power has increased in recent years.

Currently in existing wind turbines, the technique of data transfer and sharing using Ethernet protocols such as TCP / IP or UDP is a software-based method, managed by the wind turbine controller software. However, the software-based approach such as TCP / IP or UDP has the disadvantage that it is not deterministic due to, for example, network collisions or message collisions.

Document EP2410174 discloses a wind turbine in which the control units are configured to provide control redundancy. First and second turbine control units (TCU) receive monitoring data from the turbine in parallel, the first being a primary control unit and the second being a secondary control unit. If there is a defect in the primary TCU, the secondary TCU takes over the operation, based on the previously stored status information.

Summary

According to the invention there is provided a wind turbine comprising a plurality of control systems controlling different operations of the turbine connected by a communication network; a reflective memory system comprising a memory for each control system of the plurality of control systems, the plurality of control systems being coupled with the reflective memory system, wherein each memory of the control system of the plurality of control systems is accessible to all other control systems of the plurality of control systems and wherein when data is written to the memory of one control system of the plurality of control systems, the data is automatically replicated in the memories of the other control systems.

According to one embodiment, the plurality of control systems include at least one of a power control system, a blade pitch control system, and a yaw control system.

According to one embodiment, the memory for each control system comprises a dual port memory.

According to one embodiment, a control system of the plurality of control systems is configured to inform the other control systems of the plurality of control systems of the update of the data when data is written to the memory of the control system. control.

According to one embodiment, the plurality of control systems are connected via a copper or fiber optic cable.

According to one embodiment, the plurality of control systems are connected in a ring or star topology.

According to one embodiment, the plurality of control systems are synchronized.

According to one embodiment, each of the plurality of control systems includes a synchronization clock circuit, and the plurality of control systems are synchronized with the respective synchronization clock circuit of each control system.

Brief description of the drawings

In the drawings, like reference characters generally refer to the same parts throughout the different views. The drawings are not necessarily to scale, the emphasis being generally on illustration of the principles of the invention. In the description that follows, various embodiments of the invention are described with reference to the following drawings, in which:

Figure 1 illustrates a common configuration of a conventional wind turbine.

Figure 2 shows the side view of a wind turbine according to one embodiment.

ES 2 633 166 T3

Figure 3 illustrates a data sharing mechanism between two control systems within a wind turbine.

Figure 4 illustrates the data sharing mechanism between a plurality of control systems within a wind turbine in accordance with an example embodiment.

Figure 5 shows a flow chart of a data sharing method according to an example embodiment.

Figure 6 illustrates the data sharing mechanism between two control systems within a wind turbine in accordance with an example embodiment.

Detailed description

Figure 1 illustrates a common configuration of a conventional wind turbine 100. Wind turbine 100 is mounted on base 102. Wind turbine 100 includes tower 104 having a number of tower sections. A wind turbine nacelle 106 is positioned on top of tower 104. The wind turbine rotor includes a hub 108 and at least one rotor blade 110, for example three rotor blades 110.

Rotor blades 110 are connected to hub 108 which in turn connects to nacelle 106 via a low speed shaft extending outward from the front of nacelle 106. The low speed shaft typically drives a generator ( not shown) for electrical power production. The electrical energy generated is typically subsequently conditioned by a converter system (not shown), comprising a power converter, prior to delivery from a wind turbine to an electrical grid.

Figure 2 shows the side view of a wind turbine 200 according to one embodiment. Illustratively, wind turbine 200 may include a plurality of control systems (or controllers) 201, 202, 203, 204, and 205. Control systems 201-205 may be connected via a communication network. It is to be noted that the number and locations of the control systems shown in Figure 2 are not limited thereto but for illustration purposes only. Illustratively, for example, control systems 201 and 202 may comprise blade pitch control systems. Control system 203 may comprise a power control system. Control system 204 may comprise a yaw control system. Control system 205 may comprise a main control system. Wind turbine control systems 201-205 can be connected using a deterministic communication network that has consistent systems and deterministic data transfer latency. In general, a deterministic communication network ensures that data is available at a fixed rate on all nodes in the system. Wind turbine 200 may additionally include a reflective memory system (not shown). The plurality of control systems 201-205 can be coupled to the reflective memory system.

Reflective memory is a type of distributed shared memory (DSM). In general, reflective memory is a hardware-based communication mechanism for sharing data between computers and allows simultaneous reads and writes to multiple memories. Reflective memory provides a mechanism for sharing data between nodes in a distributed system. Unlike TCP / IP or UDP, reflective memory provides deterministic operation without adding software overhead. Deterministic real-time data communication network is an important prerequisite for a successful DSM implementation. The backbone of the control and communication of the wind turbine can be based on a Time Activated Protocol a type of deterministic communication network in real time.

In accordance with one embodiment, the reflective memory system in the wind turbine 200 may include a memory (eg, dual port memory) for each of the plurality of control systems 201-205 so that data can be shared among the plurality of control systems 201 to 205. For example, the memory of each plurality of control systems may have a temporary store (for example, memory in hardware) in the local control system, or another data structure in a shared system memory (for example, availability of storage space). memory in other global nodes), or cloud of data shared between control systems. In an exemplary embodiment, each control system of the plurality of control systems 201-205 can host a DSM interface that contains a fixed amount of onboard memory, which can, for example, range from kilobytes to several megabytes. This onboard memory can be used as the shared memory that can be accessed simultaneously by multiple wind turbine programs with the intention of providing communication between them or avoiding redundant copies. A control system (for example 201) of the plurality of control systems may be configured to inform (for example by interrupt) the other control systems (for example 202-205) of the plurality of control systems of the update of the data when data is written to the memory of the control system (eg 201).

Depending on the context, the wind turbine program (s) can run on a single processor or multiple separate processors. Through data sharing between the various wind turbine control systems, each control system can always have an up-to-date copy of the shared fixed memory transmitted on its highly deterministic, real-time data communication network. Each wind turbine control system that has the DSM interface can be connected via copper or fiber optic cable, in a ring or star topology with or without a network switch. When a wind turbine control system writes

ES 2 633 166 T3 data on its DSM interface, the hardware can automatically duplicate the data to the memory of all other control systems on the network. Each control system may be able to see the data almost simultaneously at the same direction offset when it was written to the source node. The transport mechanism, error checking, arbitration, etc. it can be transparent to the user.

According to one embodiment, when data is written to the memory of the control system of the plurality of control systems 201-205, the data is automatically replicated in the memories of the other control systems. For example, when data is written to the memory of the control system 201, the data can be automatically replicated in the memories of the other control systems 202 to 205. The communication network may have hardware (eg trigger lines) and software (eg interrupts) capability to perform signaling from one control system to another control system when events occur (eg data update). For example, when data is written to the memory of control system 201, the trigger line connecting to control system 201 and 203 can be configured to generate an interrupt for control system 203 that informs control system 203 of the event. data is written to the memory of the control system 201. Upon receipt of the interrupt, the control system 203 may replicate the data or read the data from the memory of the control system 201.

According to one embodiment, each control system memory of the plurality of control systems 201-205 is accessible to all other control systems of the plurality of control systems. For example, control system memory 201 may be accessible to all other control systems 202-205.

According to one embodiment, the plurality of control systems 201-205 are synchronized. In a further embodiment, each of the plurality of control systems 201-205 includes a synchronization clock circuit, and the plurality of control systems are synchronized with the respective synchronization clock circuit of each control system.

Figure 3 illustrates the data sharing mechanism between two control systems.

The reflective memory system in the wind turbine 200 shown in Figure 2 can allow synchronization of data generation and consumption among the plurality of control systems. For example, each control system can signal the others when new data has been published or some other action needs to be taken by another control system.

In one embodiment, the reflective memory system can signal to the other control system (s) through hardware interrupts whose purpose is to inform a processor (CPU) of another control system that it has occurred. an event that requires immediate service.

Interrupts can be generated locally or remotely. Reflective memory can be programmed to generate a local interrupt to inform your local CPU whenever some event has occurred (for example, a particular data location has been updated). Similarly, remote outages can be generated when one control system needs to inform another control system that an event has occurred.

Through the use of interrupts, a handshake mechanism can be created between multiple control systems to effectively synchronize data transfer.

Figure 3 shows an example execution flow using interrupts. In this example, control system A is transmitting a data stream to control system B through the reflective memory network. The greeting can be used so that control system A can inform control system B when new data is ready. As shown in Figure 3, when data is written to the memory of control system A, an interrupt is generated. The control system B waits for the interrupt, and when the control system B receives an interrupt, the control system B can read data from the control system A. After the control system B reads the data residing in the control system A, control system B can send an acknowledgment to control system A, and then wait for a next interrupt.

Figure 4 illustrates the data update mechanism between different control systems within a wind turbine.

For illustration purposes only, suppose there are three control systems in the wind turbine, i.e. a control system 401 for the wind turbine rotor, a control system 402 for the wind turbine drive train, and a 403 control system located in the wind turbine tower. Each of the control systems 401, 402, and 403 may include a local memory (for example DSM) 411 which may be volatile or non-volatile, a processor 412, a communications port 413, a timing circuit 414, and hardware input / output 415. Control systems 401 to 403 can be connected to each other in a ring or star topology. The memories 411 of the control systems 401 to 403 can be connected, for example, to a global memory 430. For example, the I / O hardware of the control system 401 can be connected to sensors or actuators for measurement of wind speed and wind direction. The input / output hardware of the

ES 2 633 166 T3 control system 402 can be connected to sensors / actuators for measurement of generator temperature, converter voltage and current. The I / O hardware of the 403 control system can be connected to sensors for the motors and encoder. The control system 403 can be a main controller in the wind turbine, and the control systems 401 and 402 can be distributed control systems in the wind turbine.

In Figure 4, for example, when data is updated in memory 411 of control system 401, control system 401 can be configured to generate an interrupt that informs control systems 402 and 403 of the data update. Control systems 402 and 403 can then read the data from local memory 411 of control system 401. Alternatively, the updated data in control system 401 can be duplicated onto global memory 430 that is accessible to control systems 402 and 403.

Figure 5 shows a flow chart 500 of a method for updating data between control systems within a wind turbine in accordance with an example embodiment.

At 501, control systems (eg distributed control nodes (DCN)) that are connected together via a common deterministic network have their operations synchronized with their built-in timing clock circuit. For example, referring to Figure 4, the control systems 401 to 403 can be connected together via a common deterministic network and have their operations synchronized with the respective integrated timing clock circuit.

At 502, the input hardware (eg, analog-to-digital converter) of the DCN 1 control system (eg, the control system 401 shown in Figure 4) acquires and digitizes a sensor signal.

On the 503, the digitized sensor signal is written to the local memory of the DCN 1 (for example DSM), which can be configured as a circular first-in-first-out (FIFO) buffer. The FIFO data storage operation can be managed and controlled by each integrated sync clock circuit of the DCN. In a distributed control node (DCN) system, a circular FIFO buffer can be implemented using a section of distributed shared memory.

At 504, the data from DCN 1 in the local FIFO is replicated to buffers of all other DCNs (for example the 402 and 403 control systems shown in Figure 4) on the same network using the synchronization method described with reference to Figure 3. That is, when the digitized sensor signal is written to the local memory of DCN 1 (for example control system 401), an interrupt is generated that informs all other DCNs (for example control systems 402 and 403) so all other DCNs can read data from DCN 1 local memory and replicate the data.

Figure 6 illustrates the data update process between control system 601 and control system 602 within a wind turbine. The 601 and 602 control systems can be connected together using a common deterministic network.

Each of the 601 and 602 control systems includes 611 input hardware (such as an analog-to-digital converter (ADC) or DI channels), 612 output hardware (such as a digital-to-analog converter (DAC) or DO channels) , a timing circuit 613, a processor 614, a local memory (eg DSM) 615, and a communications port 616. The input hardware 611 can be connected to sensors 621. The 611 input hardware can be configured to digitize analog or digital input signals. Output hardware 612 can be configured to produce analog or digital output signals to actuators or motors 622. Synchronization circuit 613 can be configured to synchronize input hardware 611 and output hardware 612. Synchronization circuits 613 in control systems 601 and 602 can also be configured to synchronize control systems 601 and 602. Digitized data can be processed in processor 614 and processor 614 can be configured to write processed data to memory 615 .

An interrupt may be generated by communication port 616 of control system 601 to communication port 616 of control system 602. Control system 602 waits for the interrupt. Upon receiving the interrupt from the control system 601, the control system 602 is configured to read data from the memory 615 of the control system 601 and then replicate the data in the local memory 615 of the control system 602.

Various embodiments provide a hardware-based method for sharing data between a wind turbine control network that includes multiple wind turbine control systems using distributed shared memory access techniques, for example reflective memory. The wind turbine control system can be connected to an external monitoring and / or control system from which it receives control commands and to which it sends data regarding the operation of the wind turbine. Said external control and / or supervision system can be a SCADA (control, supervision and data acquisition) system arranged to supervise and control the operation of the wind turbine. Typically, the external monitoring and / or control system is located remotely from the wind turbine generator so as to facilitate control and / or monitoring of the wind turbine generator operation from a remote location.

ES 2 633 166 T3

The distributed shared memory described in this document is different from other data sharing technologies such as TCP / IP or UDP that are commonly used in "relative" applications in real time, since it is a communication mechanism implemented in hardware to share data between wind turbine control systems.

Various embodiments provide a hardware-based approach, providing deterministic data flow in a distributed wind turbine control network that is comprised of multiple wind turbine control systems.

Various embodiments provide an integrated hardware- and software-based method for sharing data between monitoring and control systems in the wind turbine using distributed shared memory (reflective memory).

The wind turbine as described herein provides the ability to share data in a timely and deterministic manner between multiple wind turbine control systems while meeting the performance and determinism requirements of the entire system. Timely, deterministic data sharing is important for critical wind turbine control operations such as pitch or yaw control systems. Distributed shared memory enables deterministic synchronization between all control systems within the wind turbine through improved determinism of critical tasks over time. Hence the optimization of the wind turbine generator control operations.

Furthermore, the wind turbine as described herein provides the ability to transmit a live data stream and support large amounts of data in the wind turbine control network.

Additionally, the wind turbine as described herein ensures data delivery to ensure that critical data is not lost.

Additionally, using distributed shared memory (reflective memory), a simulation model can be divided to run simultaneously on different control systems. Splitting the processing load between two or more control systems within the wind turbine can achieve the scalability of the system and achieve optimal performance of the wind turbine generator. Input and output values ​​can be shared between individual control systems through distributed shared memory (reflective memory).

Although embodiments of the invention have been shown and described with reference to specific embodiments, it should be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention as defined in the appended claims. . The scope of the invention is thus indicated by the appended claims and all changes that fall within the meaning and range of equivalences of the claims are therefore intended to be encompassed.