Systems and methods for controlling operation of co-simulator

The collaborative simulator system with an integrated optimizer optimizes co-simulation by adjusting sub-simulator inputs and parameters during iterations, addressing the inefficiencies of traditional co-simulation methods and enhancing system performance and accuracy.

JP2024160931AActive Publication Date: 2024-11-15KK TOSHIBA
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
JP2024027658
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-02
Filing Date
2024-02-27
Publication Date
2024-11-15
Estimated Expiration
2044-02-27

AI Technical Summary

Technical Problem

Existing co-simulation methods struggle to optimize the performance of complex systems due to independent operation of sub-simulators, lacking integration of parameter changes across subsystems, and traditional global optimization failing to consider individual subsystem constraints.

Method used

A collaborative simulator system with integrated optimizer that adjusts sub-simulator inputs and parameters during iterations, using optimization algorithms to optimize system performance by influencing co-simulation dynamically.

Benefits of technology

Enhances co-simulation efficiency and accuracy by allowing real-time adjustment of sub-simulator inputs, optimizing system performance and preventing suboptimal outputs through iterative optimization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024160931000001_ABST
    Figure 2024160931000001_ABST
Patent Text Reader

Abstract

To provide a system for controlling the operation of a co-simulator comprising two or more sub-simulators 304-1, 304-2, each sub-simulator being configured to simulate the behavior of a respective sub-system of a second system.SOLUTION: In a co-simulator, outputs of sub-simulators are combined to determine a value of one or more properties of a second system. The system comprises: a controller configured to receive, at one or more time points during each iteration of simulation, results of intermediate calculations performed by one or more of the sub-simulators, and to coordinate inputs of the received results; and an optimizer 302 configured to determine, at each of the one or more time points and based on the results of the intermediate calculations, one or more elements of the co-simulator to adjust so as to optimize a process of determining the value of the one or more properties of the second system.SELECTED DRAWING: Figure 3
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] SUMMARY OF THE DISCLOSURE The embodiments described herein relate to systems and methods for controlling the operation of a co-simulator that comprises two or more sub-simulators. [Background technology]

[0002] Traditionally, reduced order models including simple look-up tables are often used for simulating complex systems with multiple subsystems. While such approaches can provide results within a reasonably short time, they cannot identify or study some transient effects or obtain accurate overall system performance. Therefore, it is essential to use the most detailed models possible to maximize the quality of the system simulation results.

[0003] As a result of the above, there is an increasing trend towards co-simulation architectures, or co-simulation. Co-simulation often refers to the joint simulation of loosely coupled stand-alone sub-simulators that can be used to model or simulate different subsystems of an overall larger system, often in different environments. Several products currently provide support for this type of system simulation, as there are many applications in the automotive and aerospace industries, for example.

[0004] When implementing co-simulation, data flow between each model is realized using defined structures that allow the co-simulator to control the overall simulation of the entire system. Co-simulation algorithms handle the time synchronization and interaction across the sub-simulators. The sub-simulators are assumed to be completely independent of each other with communication points implemented by the co-simulation controller. The sub-simulators may use different software or simulation approaches and may operate in different simulation domains such as electromagnetic, thermal or mechanical domains.

[0005] Conceptually, a sub-simulator behaves like a black box, accepting inputs from other sub-simulators, progressing forward in time with embedded solver routines to the next communication point, and finally outputting results, which may then be used again as inputs for other sub-simulators.

[0006] Although co-simulation can provide useful information about how well a system is likely to perform given a particular set of input values ​​and parameters, using co-simulation to optimize the performance of the system can be difficult. Traditionally, individual subsystems are optimized independently of each other without taking into account how changes in input values ​​and parameters of one sub-simulator may affect the performance of another sub-simulator. In other methods, a global approach may be used in which the final output from the co-simulation is fed into an optimization algorithm. However, such global optimization may not take into account constraints imposed by individual subsystems and may lead to less than optimal output from individual parts of the system.

[0007] Embodiments will now be described, by way of example only, with reference to the accompanying drawings, in which: [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 illustrates an exemplary structure of a conventional co-simulator. [Diagram 2] 2 illustrates an example sequence of operations performed by the co-simulator of FIG. 1. [Diagram 3] FIG. 2 is a diagram showing the structure of a co-simulator according to an embodiment. [Figure 4] 4 illustrates an example sequence of operations performed by the co-simulator of FIG. 3. [Diagram 5] Schematic diagram showing different types of floating structures for offshore wind turbines. [Figure 6] FIG. 6 illustrates a number of blade control parameters that may be adjusted to optimize power output from the turbine shown in FIG. 5 . [Figure 7] 1 is a schematic diagram illustrating a co-simulator for use in simulating the functioning of a wind turbine according to an embodiment. [Figure 8] 1 is a table showing a comparison of the properties of various types of optimization algorithms for use in the embodiments described herein. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] According to a first embodiment, there is provided a system for controlling the operation of a co-simulator comprising two or more sub-simulators, each sub-simulator configured to simulate the behavior of a respective subsystem of a second system, the co-simulator being configured to perform one or more iterations of a simulation in which outputs of the sub-simulators are combined to determine values ​​of one or more properties of the second system, the system comprising: a controller configured to receive, at one or more points during each iteration of the simulation, results of intermediate calculations performed by one or more of the sub-simulators, and to adjust inputs of the received results to other of the sub-simulators for use in subsequent calculations to be performed in the course of the iteration; and an optimizer configured to determine, at each of the one or more points in time, based on results of the intermediate calculations, one or more elements of the co-simulator that should be adjusted to optimize a process of determining values ​​of the one or more characteristics of the second system.

[0010] One or more sub-simulators may simulate the behavior of the respective subsystems by executing the respective models, and one or more elements of the co-simulator may comprise one or more parameters of at least one of the models.

[0011] The optimizer may be configured to adjust parameters of at least one of the models so that the model may be executed more quickly by the respective sub-simulator.

[0012] One or more elements of the co-simulator may provide intervals at which intermediate results are output by one or more sub-simulators to the controller.

[0013] The optimizer may be configured to abort a current iteration of the simulation based on determining that one or more of the intermediate results are outside a predetermined range of values.

[0014] The value of the one or more characteristics of the second system may comprise a measure of performance of the second system.

[0015] In some embodiments, for each iteration of a simulation, two or more sub-simulators may be provided with one or more input values; The optimizer may be configured to determine a revised set of input values ​​to use in the next iteration based on the estimated values ​​of the one or more characteristics of the second system.

[0016] The revised set of input values ​​may be determined using an optimization algorithm, the optimization algorithm being configured to determine a set of input values ​​that yields optimal values ​​for one or more characteristics of the second system.

[0017] The co-simulator may be configured to determine values ​​of two or more characteristics of the second system, and an optimization algorithm may be selected to determine a set of input values ​​that jointly optimize the values ​​of the two or more characteristics of the second system.

[0018] The optimizer may be configured to switch between using alternative optimization algorithms over the course of the co-simulator running multiple iterations of the simulation.

[0019] The optimizer may be configured to switch to using an alternative optimization algorithm in response to adjusting one or more elements of the co-simulator.

[0020] When switching to use an alternative optimization algorithm, the optimizer may be configured to retain the set of input values ​​determined in the most recent iteration of the simulation as inputs for the next iteration of the simulation.

[0021] The optimization algorithm may be selected based on one or more characteristics of at least one sub-simulator of the two or more sub-simulators.

[0022] The one or more sub-simulators simulate the behavior of the respective subsystems by executing the respective models. One or more elements of the co-simulator may comprise one or more parameters of at least one of the models. An optimization algorithm may be selected based on the one or more parameters.

[0023] The optimization algorithm may be selected based on a precision level or a noise level associated with one or more of the sub-simulators.

[0024] The accuracy of each sub-simulator may be defined based on the difference between the results of the calculations on the variables output by the sub-simulator and the expected values ​​for the variables in the real world.

[0025] The optimization algorithm may be selected based on the memory requirements of one or more of the sub-simulators.

[0026] The second system may be a wind mill or a wind turbine.

[0027] According to a second embodiment, there is provided a method for managing the operation of a co-simulator comprising two or more sub-simulators, each sub-simulator configured to simulate the behavior of a respective subsystem of a system, the co-simulator being configured to perform one or more iterations of a simulation in which outputs of the sub-simulators are combined to determine values ​​of one or more properties of the system, the method comprising: receiving, at one or more points during each iteration of the simulation, results of intermediate calculations performed by one or more of the sub-simulators; adjusting the input of the received results to other of the sub-simulators for use in subsequent calculations to be performed in the course of the iteration; and determining, at each of the one or more points in time, based on results of the intermediate calculations, one or more elements of the co-simulator that should be adjusted to optimize a process of determining values ​​of the one or more characteristics of the system.

[0028] According to a third embodiment, there is provided a non-transitory computer-readable storage medium comprising computer-executable instructions which, when executed by a computer, cause the computer to perform the method according to the second embodiment.

[0029] 1 shows the structure of a conventional co-simulator 101. The co-simulator 101 comprises a number of sub-simulators 104 and a co-simulation controller 103. In this embodiment, the co-simulator includes two sub-simulators 104-1 and 104-2, but in general, the co-simulator may handle any number of sub-simulators depending on the complexity of the system to be modeled and the number of subsystems that are required to be simulated.

[0030] The sub-simulator 104 may be entirely software-based and may consist of two main components: a model and a solver. A model generally comprises one or more variables that represent the characteristics of a particular subsystem being simulated. These variables may include variables that represent the physical characteristics of the subsystem, such as the physical dimensions or physical weight of the components, the operating parameters of the subsystem, such as input power, the environmental conditions in which the system is operating, such as temperature, wind speed, pressure, etc. A model may include one or more mathematical equations that can be solved by a solver to model how changes in each variable affect the overall output of the subsystem. In some implementations, a model may be used to determine how a particular variable changes over time.

[0031] In some examples, the model for a particular sub-simulator may include one or more ordinary or partial differential equations whose solutions are determined iteratively, for example using a numerical integration algorithm. In such cases, the solver may be configured to run the algorithm for a predetermined number of iterations, or until the output from the model reaches a point beyond which it stops converging further.

[0032] The sub-simulators 104-1 and 104-2 are loosely coupled to each other by communicating with the co-simulation controller 103. The co-simulation controller 103 synchronizes the co-simulation 101 by implementing communication points between the sub-simulators 104-1 and 104-2. At each communication point, one or more of the sub-simulators 104-1, 104-2 return values ​​of one or more variables to the co-simulation controller. The co-simulation controller 103 may then communicate those output values ​​to a different one of the sub-simulators 104-1, 104-2 to be used as input values ​​in a subsequent calculation. This process may be repeated a defined number of times, with the sub-simulators exchanging results of their calculations at multiple time intervals before the co-simulation reaches its final output. The results exchanged at each communication point may be considered "intermediate results" in that they are calculated by the respective sub-simulators during the course of a separate iteration of the co-simulation and are used to determine the final output of the co-simulation for that particular iteration.

[0033] The final output from the co-simulator may itself specify values ​​for one or more characteristics of the system being simulated. For example, the final output may comprise an estimate of how well the system is likely to perform under some conditions. Upon obtaining the final output for a particular iteration of the co-simulation, the co-simulation may be re-run over one or more successive iterations, with the sub-simulators initialized with different starting values ​​and parameters at the start of each iteration. In this way, it is possible to observe the effect of varying these values ​​and parameters on the final output of the system.

[0034] Still referring to Figure 1, the system may be considered to include an external optimization module, or optimizer 102. The optimizer 102 is configured to select initial starting inputs for each sub-simulator 104-1, 104-2 for each iteration of the co-simulation. To do so, the optimizer may use an optimization algorithm designed to identify input values ​​that produce the best results from the co-simulation, e.g., the algorithm may be used to determine a set of input values ​​that optimize the performance of the system, as measured by the values ​​specified in the final output.

[0035] The functionality of the co-simulator described above may be further understood with reference to FIG. 2, which shows the sequence of operations performed by the sub-simulators 104-1 and 104-2, the controller 103, and the external optimizer 102.

[0036] The process begins in steps S201 and S202 where sub-simulators 104-1 and 104-2 receive input values ​​for their sub-simulations. If it is the first iteration of the co-simulation, these input values ​​may be determined by a user, or they may be automatically selected by an optimizer.

[0037] In step S205, a first one of the sub-simulators 104-1 uses the received inputs to calculate values ​​for one or more variables in the subsimulation of that sub-simulator. The sub-simulator 104-1 may perform these calculations iteratively using a numerical method preselected based on the step size desired accuracy and / or processing time. Upon calculating values ​​for one or more variables, the first sub-simulator 104-1 outputs those values ​​to the controller. In step S207, the controller provides the received values ​​to the second sub-simulator 104-2 to be used as inputs for the calculations performed by the second sub-simulator.

[0038] The second sub-simulator 104-2 uses the input values ​​provided to it together with the values ​​it receives from the first sub-simulator 104-1 at the beginning of the simulation to calculate values ​​for one or more variables in the subsystem being simulated by the second sub-simulator 104-2. As with the first sub-simulator, the second sub-simulator may perform its calculations iteratively using a numerical method with a step size preselected based on the desired accuracy and / or processing time (step S209). Once values ​​for one or more variables are reached, the second sub-simulator 104-2 outputs those values ​​to the controller.

[0039] In step S211, the controller provides the values ​​received from the second sub-simulator to the first sub-simulator 104-1. The process continues to repeat with steps S213 to S217 replicating steps S205 to S209, respectively. Upon receiving the latest output from the second sub-simulator 104-2 in step S219, the controller can determine the final output for the simulation (while FIG. 2 shows that the first sub-simulator 104-1 and the second sub-simulator 104-2 exchange values ​​three times over the course of the overall simulation iteration, it will be appreciated that in practice the sub-simulators may exchange values ​​any number of times as determined by the nature of the system being simulated).

[0040] The output of the current iteration is obtained in step S219 and is forwarded to an (external) optimizer. The external optimizer receives the co-simulation output and performs a cost analysis to determine a revised set of input values ​​to use in the next iteration of the co-simulation. The improved set of input values ​​may be determined by using an optimization algorithm such as an evolutionary algorithm or a Bayesian algorithm. These input values ​​are provided to the first sub-simulator 104-1 and the second sub-simulator 104-2 to be used in the next iteration of the co-simulation.

[0041] In some embodiments, the co-simulation may be configured to repeat (i.e., perform successive iterations) until the output from the co-simulation stops converging further. Alternatively, the co-simulation may continue to repeat for a predetermined length of time, thereby preventing the co-simulation from running indefinitely if a convergence event cannot be reached, or from running too long if the calculation of intermediate results in each iteration takes longer than expected.

[0042] A problem with the co-simulation system shown in Figures 1 and 2 is that the optimizer cannot influence the co-simulation during the course of each iteration. In this sense, the co-simulation is treated as a "black box" by the optimizer 102, and the optimizer must wait for that particular iteration to complete before it can affect the execution of the co-simulation by changing input values.

[0043] The embodiments described herein provide a solution to the aforementioned problems by incorporating an optimizer within the co-simulation. Figure 3 shows an exemplary configuration according to the embodiments. Compared to Figure 1, it can be seen that the optimizer 302 is now incorporated into the co-simulation controller 303. Whereas previously the co-simulation behaved as a black box from the perspective of the optimizer, the current configuration allows the optimizer the ability to control the parameters of the co-simulation during each individual iteration of the co-simulation.

[0044] The functionality of the controller / optimizer shown in Figure 3 may be further understood with reference to Figure 4, which shows an example sequence of actions performed by the various components of Figure 3. Note that in Figure 4, the optimizer is combined with the controller, rather than being an external element of the system as in Figure 2.

[0045] As in the previous case, the sequence of steps starts with steps S401, S403, in which the sub-simulators 304-1, 304-2 shown in FIG. 3 receive input values ​​for their sub-simulations. If it is the first iteration of the co-simulation, these input values ​​may be determined by a user or they may be automatically selected by an optimizer. In step S405, a first one of the sub-simulators 304-1 uses the received inputs to calculate values ​​for one or more variables in its subsimulation. The sub-simulator 304-1 may perform these calculations iteratively using a numerical method with a step size preselected based on the desired accuracy and / or processing time. Having calculated values ​​for one or more variables, the first sub-simulator 304-1 outputs those values ​​as intermediate results to the controller and optimizer.

[0046] In step S407, the controller provides the intermediate results received by the controller from the first sub-simulator 304-1 to the second sub-simulator 304-2 to be used as inputs for the calculations performed by the second sub-simulator. At the same time, the optimizer decides whether to adjust one or more elements of the co-simulator to optimize the overall simulation.

[0047] The optimizer may adjust the co-simulator in one of several different ways. In one example, the optimizer may determine that a first sub-simulator is taking too long to compute intermediate results and, in doing so, is slowing down the overall co-simulation. The optimizer may address this by reconfiguring the model being run by the sub-simulator so that it issues intermediate results more quickly. For example, if the model is a model that uses an iterative method to compute results, the optimizer may reduce the extent to which the results of each iteration of the model are required to converge before being output to the controller. If the model is used to determine how the values ​​of certain variables evolve over time, the optimizer may increase the time steps used in successive iterations of the model such that a less granular, and possibly less accurate, approach is used by the sub-simulator that still reaches an output in a shorter period of time. In some embodiments, the optimizer may configure the sub-simulator to switch to an entirely different model to compute the required variables. Alternatively, or in addition, the optimizer may issue a request for more memory to be allocated to the sub-simulator in question, to increase the processing power of that sub-simulator.

[0048] In a further embodiment, the optimizer may determine that the current iteration of the co-simulation is not worth continuing based on the results received from the sub-simulators. The optimizer may determine, for example, that the results are outside an acceptable range of values, including cases where the sub-simulator does not return the actual value or values, but instead returns an error message or other output that suggests that the model cannot run successfully using that particular set of input values. The optimizer may determine that the set of input values ​​selected for the current iteration of the co-simulation would lead to results that are not practical or that do not satisfy safety regulations if those values ​​were actually implemented in the real-world system being simulated. The optimizer may determine that the values ​​calculated by the first sub-simulator are incompatible with the subsystem being simulated by the second sub-simulator. In such cases, the optimizer may issue instructions to abandon the current iteration of the co-simulation. The optimizer may further issue instructions to start a new iteration with a different set of input values.

[0049] It should be understood that the steps described above are not possible in the conventional co-simulation shown in FIG. 2 because the optimizer must wait until the entire co-simulation is complete before it has access to information about the co-simulation, and therefore changes to the behavior of the co-simulation are only possible for the next instance of the co-simulation.

[0050] Upon determining the adjustments (if any) to be made to the model, the optimizer communicates those adjustments to the involved sub-simulators, which are updated accordingly. If the optimizer determines that the current iteration should be aborted, the co-simulation is stopped or returns to steps S401, S403, and a revised set of input values ​​is selected for the sub-simulators to begin a new iteration of the co-simulation. Assuming that the current iteration is allowed to continue to run, the method proceeds to step S409, where the second sub-simulator uses the input values ​​provided to the second sub-simulator at the beginning of the simulation together with the values ​​received by the second sub-simulator from the first sub-simulator 304-1 to calculate values ​​for one or more variables in the subsystem being simulated by the second sub-simulator 304-2. As in the case of the first sub-simulator, the second sub-simulator may perform its calculations iteratively using a numerical method whose step size is preselected based on the desired accuracy and / or processing time. Once values ​​for one or more variables have been arrived at, the second sub-simulator 304-2 outputs those values ​​to the controller / optimizer.

[0051] In step S411, the controller provides the intermediate results received by the controller from the second sub-simulator 304-2 to the first sub-simulator 304-1 to be used as inputs for further calculations performed by the first sub-simulator 304-1. At the same time, based on the intermediate results received from the second sub-simulator 304-2, the optimizer decides whether to adjust one or more elements of the co-simulator to optimize the overall simulation. Here, the optimizer may perform steps similar to those described above in step S407.

[0052] Again, assuming the optimizer does not signal that the current iteration be stopped, the process repeats with steps S413 to S417 replicating steps S405 to S409, respectively. In step S419, having received the latest outputs from the second sub-simulator 304-2, the controller can determine the final outputs for the simulation (although FIG. 4 shows that the first sub-simulator 304-1 and the second sub-simulator 304-2 have exchanged values ​​three times over the course of the overall simulation iteration, it will be appreciated that in practice the sub-simulators may exchange values ​​any number of times, as determined by the nature of the system being simulated).

[0053] Thus, by incorporating the optimizer 302 within the co-simulation, the optimizer can analyze the behavior of the co-simulation during each individual iteration of the co-simulation and then make changes to the co-simulation that can help improve the performance and efficiency of the co-simulator. For example, it may be that a set of inputs is completely incompatible with the optimization of one of the output values. Rather than having to wait until the completion of the co-simulation to know this, the co-simulator of this embodiment can determine that these values ​​are inappropriate at a much earlier stage and take steps to prevent unnecessary use of computation time and power.

[0054] Still referring to FIG. 4, the output of the current iteration is obtained in step S419 and is now used by the optimizer to perform a cost analysis to determine a revised set of input values ​​to be used in the next iteration of the co-simulation.

[0055] 1 and 2, the optimizer may use one of several optimization algorithms, such as an evolutionary algorithm or a Bayesian algorithm, to determine the revised input values. The optimizer may select the most appropriate optimization algorithm based on the knowledge the optimizer has about the various subsystems and their respective simulation models. In some embodiments, the optimization algorithm may be selected based at least in part on user input.

[0056] It should be appreciated that the optimization performed in step S421 is a different process from the adjustments performed by the optimizer upon receiving intermediate results from the various sub-simulators. In contrast to those adjustments, which seek to optimize the functionality of each sub-simulator model and the interactions between them, the optimization performed in step S421 is designed to identify a set of one or more input parameters for the simulation that will yield the best results in terms of the final output from the co-simulation. For example, the co-simulator may seek to optimize one or more of the cost, performance, and stability of the system, and the output of the co-simulation may provide an indication of the extent to which these requirements are met for different sets of input values. In this regard, while the optimization algorithm used in step S421 is capable of identifying a set of input values ​​for which one or more characteristics of the system are optimized, the adjustments performed by the optimizer during each individual iteration of the co-simulation may make the process of performing each iteration more efficient and obviate unnecessary calculations by eliminating certain sets of input values ​​as being less than optimal earlier in the process.

[0057] The optimization algorithm used by the optimizer may change dynamically during the course of the co-simulation. For example, some of the subsystems may use a simple simulation model during normal operating conditions. However, in extreme conditions, a different / more accurate model may be required, which will change the run time for each iteration of the co-simulation. For example, if a more accurate model that takes longer to calculate is used for one of the subsystems, the optimization algorithm used in step S421 may be adjusted so that it requires fewer evaluations, i.e., fewer iterations of the overall co-simulation. In such cases, the co-simulator may transfer the current state of the old optimization algorithm to the new optimization algorithm to avoid a "cold start".

[0058] In some instances, some subsystems may use less accurate simulation models than other subsystems, and if the co-simulator knows this, it may choose a different optimization algorithm or change the parameters of the current optimization algorithm. For example, when hardware-in-the-loop is present, there may be measurement noise that can be quantified. In other cases, as explained above, some sub-simulations may require a lot of memory to process a particular set of input values. To address this, the optimizer may switch to an alternative optimization method or change the parameters of the current optimization method.

[0059] The embodiments described herein may be utilized for design automation purposes. When designing a product, co-simulation may determine optimal configurations of design parameters of one or more of the subsystems in order to optimize one or more of the system's characteristics and provide those configurations to a user for manufacturing. Embodiments may also be created to determine how an existing system behaves under a set of different circumstances in order to optimize the operation of the existing system.

[0060] An exemplary co-simulator implementing the concepts described above will now be described with reference to Figures 5 to 7. In this example, the co-simulator is used to optimize the design and / or operation of an offshore wind turbine. Figure 5 shows different types of floating structures 501-1, 501-2, 501-3, 501-4 for an offshore wind turbine, while Figure 6 shows a number of blade control parameters that may be adjusted to optimize the power output from the turbine. Figure 7 shows a schematic diagram of a co-simulator for use in simulating the functioning of a wind turbine.

[0061] 7, a co-simulator 701 comprises a co-simulation controller 701 and optimizer 703, as well as three independent sub-simulators 704-1, 704-2, and 704-3 representing the subsystems of an offshore turbine. In this embodiment, sub-simulator 704-1 is used to simulate the power generation equipment, sub-simulator 704-2 simulates the blades of the turbine, and sub-simulator 704-3 simulates the physical structure on which the turbine is based.

[0062] The sub-simulator 704-1 is primarily concerned with simulating the power generation by the generator. During each co-simulation, the sub-simulator 704-1 receives input values ​​from the co-simulation controller that define the blade rotation speed and the system reference power. At the beginning of the co-simulation, these values ​​may be initially set by the user or they may be automatically selected by the co-simulation controller / optimizer. The generator sub-simulator 704-1 also includes a torque controller for the purpose of modeling how torque changes affect power generation. Since torque changes also affect the turbine rotation speed, the rotation speed may also be an output at the communication point for use in the other sub-simulators along with the determined power output.

[0063] The blade sub-simulator 704-2 may include a blade pitch controller for the purpose of modeling the effect of varying the blade pitch on the blade behavior, which may be output at the communication points in the form of one or more output values ​​representing, for example, the determined rotational speed, the blade pitch angle, and the degree of blade bending.

[0064] A sub-simulator 704-3 representing an offshore structure receives as input data representing wind speed, wave spectrum, and blade pitch angle. The sub-simulator 704-3 uses these as inputs to calculate a six degree movement of the structure in time. This six degree movement may then be provided to a co-simulation controller at a communication point.

[0065] As an example of an exchange of input / output values ​​between the sub-simulators, as the co-simulation progresses, the offshore structure simulator 704-3 receives input values ​​for the blade pitch angle from the blade sub-simulator 704-2, again via the co-simulation controller. The blade sub-simulator 704-2 calculates the blade pitch angle as an intermediate result during each iteration of the co-simulation. The offshore structure sub-simulator 704-3 then uses the values ​​received from the blade sub-simulator 704-2 to calculate the 6 degree movement of the offshore structure.

[0066] As described above, upon receiving the blade pitch angles from the blade sub-simulator, the controller / optimizer may evaluate whether the calculated pitch angles are within acceptable limits and further whether the blade sub-simulator is able to output the pitch angle calculations within an acceptable time frame. If one or another of these criteria is not met, the optimizer may issue instructions to adjust one or more parameters of the blade sub-simulator and / or terminate the current iteration of the co-simulation and then initialize a new iteration of the co-simulation with a new set of starting values ​​for a different sub-simulator.

[0067] It will be appreciated that some of the inputs shown in Fig. 7, such as wind speed and wave spectrum, are operating conditions for the system and only affect the system, and changes in these operating conditions may not be calculated by the sub-simulators. These operating conditions may be set as constants by the co-simulation controller, or may be varied over time to simulate the response of the system to changing conditions. Other inputs, such as rotation speed and the amount of power generated, comprise variables whose values ​​may be calculated by one or more of the sub-simulators. The behavior of these variables may be subject to simulation by multiple sub-simulators, therefore requiring communication between the sub-simulators via the co-simulation controller to ensure uniformity of the behavior of the variables between the different subsystems.

[0068] The co-simulation may be configured such that the power generated and the stability of the floating platform are specified as outputs for optimization. The power generated is the output of the generator sub-simulator, and the stability may be determined from the output of the 6 degree motion of the offshore structure sub-simulator. It may not be possible to optimize most of these outputs simultaneously, for example, there may be configurations in which power generation is maximized but the structure is unstable, leaving the turbine vulnerable to changes in wind speed or wave spectrum. Similarly, there may be different configurations in which the structure is very stable, but as a result, power generation is limited. In this case, a multi-objective optimization algorithm may be used to find a configuration for the wind turbine structure that achieves an optimal trade-off between these outputs.

[0069] Two exemplary optimization algorithms are Nondominated Sorted Genetic Algorithm (NSGA) II and Thompson Sampling Efficient Multiobjective Optimisation (TSEMO). NSGA-II is an example of an evolutionary algorithm and TSEMO is an example of a Bayesian algorithm. Both algorithms have different strengths and weaknesses that the optimization module can be aware of when selecting an algorithm to use. Figure 8 shows a table summarizing the properties of these two algorithms.

[0070] In most cases, NSGA-II exhibits good performance, a large variety of solutions, and fast computation speed. However, NSGA-II requires the performance of multiple function evaluations and has little ability to deal with errors in calculation or measurement. In comparison, TSEMO tends to have less diverse solutions and tends to be slower to compute, but requires fewer function evaluations and can incorporate known imprecision in the optimization. It can therefore be seen that co-simulation may have reasons to use both algorithms depending on the situation.

[0071] During optimization, the cost function needs to be evaluated multiple times with different input values. At the beginning of the optimization, the sub-simulation may be able to compute relatively quickly and with relatively little noise, since the sub-simulation may be operating within what may be considered normal operating conditions. However, as the simulation progresses, the operating conditions may approach the boundaries of normal operating conditions and may start to simulate abnormal operating conditions. As explained in the previous paragraph, this may require the use of a different model for the simulation, such as a more accurate model.

[0072] In this scenario, it may be most efficient to start the optimization using NSGA-II and only switch to TSEMO when the evaluation time of the subsimulation becomes more extensive. To preserve the NSGA-II optimization process, the current information acquired by NSGA-II can be converted into a format used by TSEMO. This may involve, for example, converting the population members currently present in NSGA-II into a TSEMO sampling from which a probability distribution can be constructed. Since members of the Bayesian optimization family of algorithms such as TSEMO are designed for computationally expensive systems, this may prove to be more computationally efficient compared to continuing the NSGA-II function evaluation on the subsimulator.

[0073] Another factor that may affect the selection of the optimization algorithm is the accuracy level of the sub-simulator. As explained in the previous paragraph, inaccuracies such as measurement noise may arise in the use of hardware in the sub-simulator or in errors or approximations present in the models used in the sub-simulator. Evolutionary algorithms assume that the function produces error-free results and therefore have difficulty dealing with inaccuracies. In contrast, Bayesian optimization algorithms can deal with inaccuracies because the variance of the prior probability distribution can be adjusted to incorporate the inaccuracies. Thus, if the optimization module is aware that one of the sub-simulators is operating with some level of inaccuracies, such as measurement noise, the optimization module may select a Bayesian method such as TSEMO to take this into account.

[0074] Implementations of the subject matter and operations described herein, including the structures disclosed herein and their structural equivalents, can be realized in digital electronic circuitry, or in computer software, firmware, or hardware, or in one or more combinations of these. Implementations of the subject matter described herein can be realized using one or more computer programs, i.e., one or more modules of computer program instructions encoded on a computer storage medium to be executed by or to control the operation of a data processing device. Alternatively, or in addition, the program instructions can be encoded on an artificially generated propagated signal, e.g., a machine-generated electrical, optical, or electromagnetic signal, generated to encode information to be transmitted to an appropriate receiving device to be executed by the data processing device. The computer storage medium can be, or be included in, a computer-readable storage device, a computer-readable storage substrate, a random access memory array or memory device, or a serial access memory array or memory device, or a combination of one or more of these. Additionally, a computer storage medium is not a propagated signal, but a computer storage medium can be a source or destination of computer program instructions encoded in an artificially generated propagated signal. Also, a computer storage medium can be or be contained within one or more separate physical components or physical media (e.g., multiple CDs, disks, or other storage devices).

[0075] Although several embodiments have been described, these embodiments are presented merely as examples and are not intended to limit the scope of the present invention. Indeed, the novel methods, devices, and systems described herein may be embodied in various forms, and various omissions, substitutions, and modifications may be made to the forms of the methods and systems described herein without departing from the spirit of the present invention. The appended claims and their equivalents are intended to cover such forms or modifications that fall within the scope and spirit of the present invention.

Claims

1. 1. A system for controlling operation of a co-simulator comprising two or more sub-simulators, each sub-simulator configured to simulate the behavior of a respective subsystem of a second system, comprising: The co-simulator is configured to perform one or more iterations of a simulation in which outputs of the two or more sub-simulators are combined to determine values ​​of one or more properties of the second system, the system comprising: a controller configured to receive results of intermediate calculations performed by one or more of the two or more sub-simulators at one or more points during an iteration of the simulation, and to adjust inputs of the received results to other of the two or more sub-simulators for use in subsequent calculations to be performed during the course of the iteration; an optimizer configured to determine one or more elements of the co-simulator at each of the one or more time points based on the multiple results of the multiple intermediate calculations; A system comprising:

2. 2. The system of claim 1, wherein the one or more sub-simulators simulate the behavior of the respective plurality of subsystems by executing a respective plurality of models, and the one or more elements of the co-simulator comprise one or more parameters of at least one of the plurality of models.

3. The system of claim 2 , wherein the optimizer is configured to tune the one or more parameters of at least one model of the plurality of models.

4. The system of claim 1 , wherein the one or more elements of the co-simulator include a time interval during which the results of the intermediate calculations are output by the one or more sub-simulators to the controller.

5. The system of claim 1 , wherein the optimizer is configured to abort an iteration of the simulation if one or more of the results of the intermediate calculations are outside a predetermined range of values.

6. The system of claim 1 , wherein the values ​​of one or more characteristics of the second system comprise a measure of performance of the second system.

7. For each iteration of the simulation, one or more input values ​​are provided to the two or more sub-simulators; 2. The system of claim 1, wherein the optimizer is configured to determine a revised set of input values ​​to use in a next iteration based on the estimated values ​​of the one or more characteristics of the second system.

8. 8. The system of claim 7, wherein the revised set of input values ​​is determined using an optimization algorithm, the optimization algorithm determining a set of input values ​​that results in optimal values ​​for the one or more characteristics of the second system.

9. 9. The system of claim 8, wherein the co-simulator is configured to determine a plurality of values ​​of two or more characteristics of the second system, and the optimization algorithm is selected to determine a set of input values ​​that jointly optimize the plurality of values ​​of the two or more characteristics of the second system.

10. The system of claim 8 , wherein the optimizer is configured to switch between using alternative optimization algorithms over the course of the co-simulator performing multiple iterations of the simulation.

11. The system of claim 10 , wherein the optimizer is configured to switch to using the alternative optimization algorithm in response to adjusting the one or more elements of the co-simulator.

12. 11. The system of claim 10, wherein upon switching to use the alternative optimization algorithm, the optimizer is configured to retain the set of input values ​​determined in a most recent iteration of the simulation as inputs for a next iteration of the simulation.

13. The system of claim 8 , wherein the optimization algorithm is selected based on one or more characteristics of at least one sub-simulator of the two or more sub-simulators.

14. the one or more sub-simulators simulate the behavior of the respective plurality of subsystems by executing a respective plurality of models, the one or more elements of the co-simulator comprising one or more parameters of at least one model of the plurality of models; The system of claim 13 , wherein the optimization algorithm is selected based on the one or more parameters.

15. The system of claim 13 , wherein the optimization algorithm is selected based on an accuracy or noise level associated with one or more of the two or more sub-simulators.

16. 16. The system of claim 15, wherein the accuracy of each sub-simulator is defined based on a difference between results of calculations on variables output by the sub-simulator and expected values ​​for the variables in the real world.

17. The system of claim 13 , wherein the optimization algorithm is selected based on memory requirements of the one or more sub-simulators of the two or more sub-simulators.

18. The system of claim 1 , wherein the second system is a wind turbine.

19. 1. A method for managing the operation of a co-simulator comprising two or more sub-simulators, each sub-simulator configured to simulate the behavior of a respective subsystem of a system, comprising the steps of: The co-simulator is configured to perform one or more iterations of a simulation in which outputs of two or more sub-simulators are combined to determine values ​​of one or more properties of the system, and the method includes: receiving results of intermediate calculations performed by one or more of the two or more sub-simulators at one or more points during an iteration of the simulation; adjusting input of the received results to other sub-simulators of the two or more sub-simulators for use in subsequent calculations to be performed in the course of the iteration; and determining one or more elements of the co-simulator at each of the one or more time points based on the multiple results of the multiple intermediate calculations; A method for providing the above.

20. 1. A program for causing a computer to execute a method for managing operation of a co-simulator comprising two or more sub-simulators, each sub-simulator being configured to simulate the behavior of a respective subsystem of a system, the method comprising: The co-simulator is configured to perform one or more iterations of a simulation in which outputs of two or more sub-simulators are combined to determine values ​​of one or more properties of the system, and the method includes: receiving results of intermediate calculations performed by one or more of the two or more sub-simulators at one or more points during an iteration of the simulation; adjusting input of the received results to other sub-simulators of the two or more sub-simulators for use in subsequent calculations to be performed in the course of the iteration; and determining one or more elements of the co-simulator at each of the one or more time points based on the multiple results of the multiple intermediate calculations; A program for causing a computer to execute the program.

Citation Information

Patent Citations

  • Method and device for previous stage co-simulation

    JP2020004399A

  • Systems and methods for automatically realizing models for co-simulation

    US20190370420A1

  • Co-simulation computer system, verification system for embedded systems, and verification method for embedded systems

    WO2014038030A1

  • Co-simulation, computer system

    WO2022128478A1