EP4655854A1

EP4655854A1 - Methods of controlling distributed energy provision systems

Info

Publication number: EP4655854A1
Application number: EP24706779.6A
Authority: EP
Inventors: Steve WATTAM; Nick GOOD
Original assignee: Kraken Technologies Ltd
Current assignee: Kraken Technologies Ltd
Priority date: 2023-01-23
Filing date: 2024-01-19
Publication date: 2025-12-03
Also published as: GB202300971D0; GB2626531B; JP2026503526A; GB2626531A; WO2024156987A1; AU2024211313A1

Abstract

The present disclosure relates to methods of controlling a distributed energy provision system, the distributed energy provision system comprising a plurality of energy provision devices and a control module or system configured to control operation of the plurality of energy provision devices, the distributed energy provision system being arranged to deliver electrical power to a power grid. An exemplary method comprises: inputting first data into an optimisation algorithm to determine a power delivery parameter for each of one or more of the plurality of energy provision devices; sending a power delivery instruction to the one or more energy provision devices to operate the one or more energy provision devices for power delivery based on the respective power delivery parameter; for each of the one or more energy provision devices: obtaining second data from metering the energy provision devices during power delivery; inputting the second data into a device behaviour model of the energy provision device; estimating, using the device behaviour model, one or more device parameters of the energy provision device; and updating the optimisation algorithm using the one or more estimated device parameters respective of each of the one or more energy provision devices.

Description

METHODS OF CONTROLLING DISTRIBUTED ENERGY PROVISION SYSTEMS

FIELD OF THE INVENTION

The present disclosure relates to operation of distributed energy provision systems, such as virtual power plants, comprising a plurality of energy provision devices to deliver electrical power to a power grid.

BACKGROUND

Optimisation for distributed energy provision systems, such as (but not limited to) virtual power plants (VPP), generally relies on models of individual energy provision or generation devices and apparatuses, such as medium-scale power generating units, micro combined heat and power units, natural gas-fired reciprocating engines, small-scale wind power plants, photovoltaics units, hydroelectricity plants, biomass power plants, backup generators, energy storage systems, etc. The models simulate the way individual energy provision devices are connected, e.g. to distributed energy provision system, to one or more sources of electrical power, to other energy provision system, etc., and the way energy markets interact with them. These models may be encoded into an optimisation model used by the distributed energy provision system for controlling the operation of individual or groups of energy provision devices. Such models typically have a fixed formulation and is not adaptive to changing conditions that may affect the performance of individual energy provision devices, e.g. device storage and/or power delivery capacity, weather conditions such as temperature and humidity that could affect the performance of the energy provision devices, changes in operation efficiencies due for example to length of use, etc.

Device parameters specified by manufacturers to describe different operational aspects of an energy provision device, such as a range of state-of- charge (SoC), a range of operation power, a range of operation temperature, etc., often lack accuracy and/or specificity, and these device parameters may change over time. Additionally, many of these device parameters can vary depending on factors such as operation conditions (e.g. temperature), device power and/or SoC, device past behaviour, etc. As such, device parameters often cannot be specified accurately based on laboratory testing by the manufacturers. As a result, conventional models used for optimising the operation of energy provision devices participating in a distributed energy provision system may not be an accurate representation of real-life operations, even when such models have initially been validated through factory testing.

It is therefore desirable to provide improved methods for controlling the operation of energy provision devices in a disturbed energy provision system.

SUMMARY OF THE INVENTION

In view of the foregoing, an aspect of the present technology provides a computer-implemented method of controlling a distributed energy provision system, the distributed energy provision system comprising a plurality of energy provision devices and a control module or system configured to control operation of the plurality of energy provision devices, the distributed energy provision system being arranged to deliver electrical power to a power grid, the method comprising : inputting first data into an optimisation algorithm to determine a power delivery parameter for each of one or more of the plurality of energy provision devices; sending a power delivery instruction to each of the one or more energy provision devices to operate the one or more energy provision devices for power delivery based on the respective power delivery parameter; for each of the one or more energy provision devices: obtaining second data from metering the energy provision devices during power delivery; inputting the second data into a device behaviour model of the energy provision device; estimating, using the device behaviour model, one or more device parameters of the energy provision device; and updating the optimisation algorithm using the one or more estimated device parameters respective of each of the one or more energy provision devices.

Embodiments of the present technology provides methods for controlling the operation of a plurality of energy provision devices/systems in a distributed energy provision system (e.g. a virtual power plant) that is capable of autonomous and adaptive modification that enables changes in operating conditions to be updated. In particular, a control algorithm such as an optimisation algorithm (e.g. but not limited to a mixed-integer linear programming MILP algorithm) may be used to generate control instructions to each one or each group of participating energy provision devices for power delivery. Inputs for the control algorithm may for example be obtained from device behaviour models that each describes the operation behaviour of one (or a group of) energy provision device, and the device behaviour models or the control algorithm, according to embodiments, are initialised using first data that may, for example, include manufacturer-specified operating parameters for the plurality of energy provision devices, and power delivery parameters (e.g. rate of delivery, total energy to be delivered, etc.) in relation to the operation of the plurality of energy provision devices are determined by the control algorithm to generate instructions for the participating energy provision devices (e.g. separate instruction for each energy provision devices or a common instruction to each group of energy provision devices). The instructions are then sent to the participating energy provision devices to operate the energy provision devices for power delivery based on the respective power delivery parameters. While an energy provision device (or a group of energy provision devices) is being operated for power delivery, the device is independently metered (e.g. measurements of various parameters are taken) to obtain a set of second data that may, for example, include operating power, operating temperature, total energy delivered, etc. The obtained second data is fed into the device behaviour model of the associated energy provision device (or group of energy provision devices), and the updated device behaviour model can then be interrogated or analysed to estimate an updated set of device parameters for that energy provision device. Updated sets of device parameters for each (or each group of) participating energy provision devices are then passed to the control algorithm to update the control algorithm. In doing so, the device behaviour models, and in turn the control/optimisation algorithm that generates operating instructions for the energy provision devices, can be continually updated for changes in operating conditions, environmental conditions, deteriorations in the devices or other instruments, as well as socio-economic factors such as market fluctuations. In some embodiments, the method may further comprise storing the second data in an observation database. Through storing the second data in a database, it is possible to analyse past behaviours and any trends of the energy provision devices, which may further improve the accuracy of the device behaviour models and in turn the optimisation algorithm.

In some embodiments, the method may further comprise obtaining respective past second data for each of the one or more energy provision devices stored on the observation database and inputting the respective past second data into the respective device behaviour model for the one or more energy provision devices. Herein, "past second data" refers to any second data obtained from the energy provision devices from operation cycles preceding the current operation cycle (time period) for which the devices are being metered. This may include previous cycles on the same day or different days, and over any time scale as desired, and thresholds may be set, if desired, to limit the amount of past second data to within a relevant period.

In some embodiments, the first data may comprise manufacturer-specified device parameters for the one or more energy provision devices such as operation power, maximum/minimum state of charge, operation temperature, efficiency curve, response time, environmental parameters such as ambient temperature, humidity, or operational parameters such as operation costs, or any combination thereof.

In some embodiments, the second data may comprise device parameters obtained through metering including operation power, operation temperature, state of charge, device configuration changes, or any combination thereof.

In some embodiments, the one or more device parameters of a respective energy provision device may comprise an operation efficiency and/or a maximum operation power at a given state of charge.

In some embodiments, the device behaviour model of a respective energy provision device may be configured to simulate operation of the respective energy provision device based on the second data obtained from the energy provision device, the meter associated with the energy provision device, and the power delivery instruction associated with the energy provision device. For example, the device behaviour model may comprise a set of rules that define the relationships amongst the operation parameters of a device. In an exemplary implementation, these relationships may be defined using one or more linear equations. However, non-linear equations and other forms of constraint are of course possible. Such device behaviour models may for example be configured based on one or more physical processes within the device (e.g. chemical battery processes, inputting/outputting power of electrical components such as inverters and wiring, etc.), or an abstract representation of observed data based on bench testing. In the former, for example, a mapping between a set of observed device parameters and a set of model rules is defined. In the latter, for example, a model is an abstract representation of a set of data and so any changes to the observed data means rebuilding the model. In some embodiments, the set of rules that defines a device model may be determined automatically (without or with minimal human intervention) using a machine learning algorithm.

In some embodiments, the power delivery instruction may comprise an operation start time, an operation end time, an operation time period, an operation power at a given time, an average operation power, an overall energy to be delivered, or any combination thereof. According to these embodiments, a power delivery instruction represents an intended behaviour for the associated device, and may for example be specified intensively as actions to be performed by the device (set power to level x for y minutes) or ostensively as conditions to be met (deliver x amount of energy over the next y minutes). Different devices and/or control systems may adopt different approaches or a combination of both.

In some embodiments, the optimisation algorithm may be a solution to a mixed-integer linear programming problem. Other examples of an optimisation algorithm may include nonlinear optimisation such as gradient descent or quadratic programming, large neural net, hybrid models e.g. based on classical statistical or Bayesian inference, etc.

Another aspect of the present technology provides a computer-readable medium comprising machine-readable code, which, when executed by a processor, causes the processor to perform the method as described above. A further aspect of the present technology provides a control module arranged to operate a plurality of energy provision devices for a distributed energy provision system to deliver electrical power to a power grid, the control module comprising one or more processors configured to execute machine- readable code stored on memory to perform the method as described above.

A yet further aspect of the present technology provides a distributed energy provision system arranged to deliver electrical power to a power grid, comprising : a plurality of energy provision devices each configured to output electrical power; and a control module configured to control operation of the plurality of energy provision devices, the control module comprising memory configured to store machine-readable code and one or more processors configured to execute the machine-readable code, wherein execution of the machine-readable code causes the one or more processor to: input first data into an optimisation algorithm to determine a power delivery parameter for each of one or more of the plurality of energy provision devices; send a power delivery instruction to the one or more energy provision devices to operate the one or more energy provision devices for power delivery based on the respective power delivery parameter; for each of the one or more energy provision devices: obtain second data from metering the energy provision devices during power delivery; input the second data into a device behaviour model of the energy provision device; estimate, using the device behaviour model, one or more device parameters of the energy provision device; update the optimisation algorithm using the one or more estimated device parameters respective of each of the one or more energy provision devices.

In some embodiments, the distributed energy provision system may be a virtual power plant.

In some embodiments, the plurality of energy provision devices may comprise one or more medium-scale power generating units, one or more micro combined heat and power units, one or more natural gas-fired reciprocating engines, one or more small-scale wind power plants, one or more photovoltaics units, one or more hydroelectricity plants, one or more biomass power plants, one or more backup generators, one or more energy storage systems and devices, or any combination thereof. Implementations of the present technology each have at least one of the above-mentioned objects and/or aspects, but do not necessarily have all of them. It should be understood that some aspects of the present technology that have resulted from attempting to attain the above-mentioned object may not satisfy this object and/or may satisfy other objects not specifically recited herein.

Additional and/or alternative features, aspects and advantages of implementations of the present technology will become apparent from the following description, the accompanying drawings and the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments will now be described, with reference to the accompanying drawings, in which :

FIG. 1 shows schematically an exemplary distributed energy provision system arranged to provide energy to a power grid; and

FIG. 2 shows an exemplary method of controlling a distributed energy provision system.

DETAILED DESCRIPTION

An exemplary distributed energy provision system 100, such as a virtual power plant, is schematically shown in FIG. 1, which is arranged to provide electricity to a power grid 190.

In the present embodiment, the distributed energy provision system 100 comprises a plurality of energy provision devices and systems and a control module or control system 110 configured to control the operation of the plurality of energy provision devices and systems. The plurality of energy provision devices and systems includes, in the present embodiment, one or more solar panels or solar energy farm 120, one or more domestic or commercial electric vehicle batteries 130, one or more solar panels 140 importing energy into one or more energy storage devices or systems 150, and one or more wind turbines 160; however, other energy provision devices and systems and/or energy generating devices and systems are also possible.

The plurality of energy provision devises and systems 120-160 communicate with a control module or control system 110 via network 170 for receiving instructions from the control module/system 110 and to enable the control module/system 110 to obtain measurements and observations with respect to the operation of the energy provision devises and systems 120-160, e.g. through receiving meter readings. The control module/system 110 communicates with the power grid 190 to coordinate power delivery, and the plurality of energy provision devises and systems 120-160 are separately connected to the power grid 190 to deliver electrical power to the grid 190 under instructions from the control module/system 110.

In conventional distributed energy provision systems, operation of the plurality of energy provision devices is controlled by static device models that each describes the operation behaviour of a respective energy provision device. An energy provision device is bench tested against a set of criteria in a laboratory or standardised setting, e.g. as part of quality assurance during manufacturing, to measure a set of device parameters (e.g. capacity, efficiency, etc.) for the energy provision device. These manufacturer-specified device parameters are then input to a corresponding device model to determine the operation parameters (e.g. power delivery rate, operation time, total energy delivered, etc.) that are input to an optimisation algorithm to generate control output or instructions for controlling the operation of the energy provision device. Conventionally, such device models are either not updated (therefore static), or only updated manually by human operators when device parameters have noticeably changed, and are identified, for example, by post-hoc analyses of operation data. In some cases, secondary algorithms may be used in addition to the primary optimisation algorithm in order to compensate for inaccuracies in the primary optimisation algorithm due to inaccurate or outdated device models.

Device parameters that determine how an energy provision device operates, such as maximum output, speed of response, range of state of charge, efficiency, etc. can vary depending on many factors, and the Applicant has recognised that it is not feasible to test all possible combinations of parameters and parameter values and take into account of how the parameters may evolve over time ahead of the device being deployed. Moreover, the site at which the device is deployed may also has an impact on the device parameters; as such, it may not be possible to determine these device parameters accurately before deployment, and it may not be possible to reuse device parameters determined for one site at a different site. Some parameters, for example the amount of energy used by a device during service, can vary as a result of gradual changes in market forces, which may also have an impact on the operation of the device. Examples of changes in market forces may include an increase in availability if batteries in a market which provides additional capacity and drives usage down, or an introduction of a second market that absorbs a proportion of the activity and throughput of the devices normally participating in the original market.

Embodiments of the present technology therefore provide improved methods of controlling a plurality of energy provision devices/systems that are participating in a distributed energy provision system, such as a virtual power plant.

FIG. 2 shows schematically an exemplary control mechanism, according to embodiments of the present technology, for controlling the operation of a plurality of energy provision devices or systems in a distributed energy provision system such as the system 100 in FIG. 1. One or more stages of the control mechanism may for example be performed by the control module/system 110.

In the present embodiment, the method begins with a control algorithm such as an optimisation algorithm (e.g. an MILP algorithm or model) 200 being initialised by inputting (manually or autonomously) first data and operational constraints such as market prices 220. In some embodiments, the first data may include default, manufacturer-specified or laboratory-tested device parameters for each energy provision device, such as maximum power or energy output, average operation power, maximum/minimum or range of state of charge, operation temperature, environmental parameters such as ambient temperature, humidity, or operational parameters such as operation costs, or any combinations.

Depending on what first data is available and the structure of the control algorithm 200, the control algorithm 200 may use the received first data as input to determine appropriate operation parameters (e.g. operation power, energy output) for power delivery by the plurality of energy provision devices, or the first data may first be input into device behaviour models that each describe the operation behaviour of a respective one of the plurality of energy provision devices in order to determine the appropriate operation parameters for the plurality of energy provision devices. Then, based on the determined operation parameters, or power delivery parameters, the control algorithm 200 generates operation instructions 230 for the plurality of energy provision devices for power delivery. In some embodiments, instructions may be specific to each energy provision devices based on its respective power delivery parameters. In other embodiments, the same instruction or same set of instructions may be generated for all of the plurality of energy provision devices or a group of energy provision devices, for example if they have the same, similar, or inter-dependent device parameters or power delivery parameters.

The instructions 230 generated by the control algorithm 200 are sent to the corresponding energy provision devices to operate the plurality of energy provision devices for power deliver 240. In some embodiments, a power delivery instruction 230 may specify an operation start time, an operation end time, an operation time period, an operation power, optionally with respect to a series of timesteps, an average operation power output, an average or overall energy output, or any combinations.

The plurality of energy provision devices is then operated according to the instructions 230 they respectively received to deliver power, 240, as part of the distributed energy provision system. During power delivery, each energy provision device or each group of energy provision devices is independently metered 250 e.g. by the control module/system 110 to obtain second data. In some embodiments, the second data may include any measurements that can be taken in relation to the operation of the respective energy provision device, for example operation power output, operation temperature, state of charge, etc., and operational or environmental conditions may also be taken into account. The metering 250 of individual energy provision devices generate a set of observations 260, or second data, for the respective device, and the set of observations 260 can be stored in a database 270 either for independent analyses or for use in updating device behaviour models. In the present embodiment, the second data obtained from metering the plurality of energy provision devices is input into a respective device behaviour model 280 for each energy provision device. The device behaviour model 280 of the respective energy provision device then uses the second data of the device to simulate operation of the energy provision device based on the second data obtained from the device, the meter associated with the device, and the power delivery instruction associated with the device.

In some embodiments, the device behaviour model 280 of an energy provision device may obtain and use as inputs past second data stored on the observations database 270 from observations made on the energy provision device during previous cycles of operation. Thus, in these embodiments, the device behaviour model 280 can be updated using not only observations made during the current cycle of operations, but also make use of observations made during past cycles of operations such as multiple values obtained for one or more parameters, average values or trends for one or more parameters, etc. In doing so, the updated device behaviour model is able to more accurately reflect real-time behaviour of the energy provision device.

The device behaviour model 280 of each energy provision device can then be interrogated and analysed to obtain (updated) estimates of various parameters 290 associated with the energy provision device. For example, regression techniques (e.g. least square (linear or nonlinear), support vector regression, etc.) may be used to model the relationship between measurements and one or more variables. The updated device parameters 290 are then used to update the control algorithm 200, which uses the updated parameters 290 to determine up-to-date power delivery parameters for each energy provision device and generate power delivery instructions for each energy provision device with improved precision.

Thus, the present approach first begins with known optimisation techniques to determine various control components. Then measurements are taken during operation to update models that describe respective participating devices such that the updated models can be interpreted (e.g. using regression techniques) to obtain updated parameters that reflect more accurately the current state of the devices, for example caused by degradation of the devices over time. The updated parameters may then be used as input to the initial optimisation algorithm or model to output up-to-date instructions to operate the energy provision devices with improved performance and efficiency. The potential improvements in accuracy in modelling the behaviour of individual (or individual groups of) energy provision devices facilitate an improvement in power delivery, which can lead to an increase in revenue derived from providing the service, and reduce the likelihood of damages to the devices by operating the devices within their actual limits. The present approach advantageously improves model accuracy compared to using only laboratory-tested parameters, does not require manual input or intervention from human operators to periodically update the models, and is more efficient over periodic laboratory testing of devices.

According to present embodiments, in a distributed energy provision system comprising a plurality of energy provision devices, the plurality of energy provision devices is controlled by respective models that describe the behaviour of the devices. Techniques describe herein enable an integration of behavioural observations obtained from the plurality of energy provision devices during power delivery, e.g. through independent metering of the devices, into the models that control them through the above described mechanism, thus optimisation of the system may be performed according to current conditions. As such, techniques described herein improve the accuracy of the models without requiring formal testing and manual development by human operators, which in turn improve the effectiveness of the control or optimisation algorithm used for controlling the plurality of energy provision devices. Through improving the accuracy of the models, it is possible to reduce the frequency or likelihood of devices being scheduled to behave in ways that may result in putting unacceptable stress on the devices, thereby potentially extending the operation lifespan of the devices.

One or more aspects of the present technology may be implemented as one or more machine learning algorithms (MLAs). For example, the regression and/or optimisation approaches described above, the estimation of device parameters using the associated device model and/or the smoothing of the results generated by the optimization algorithm, e.g. an MILP algorithm, may be implemented through one or more suitable MLAs. It should be understood that different types of MLAs having different structures or topologies may be used for various tasks. However, it should be noted that the use of an MLA in embodiment(s) of the present technology is a non-limiting example of implementing the present technology, and the use of an MLA is not essential.

Broadly speaking, there are three types of MLAs: supervised learning-based MLAs, unsupervised learning-based MLAs, and reinforcement learning-based MLAs. Supervised learning MLA process is based on a target - outcome variable (or dependent variable), which is to be predicted from a given set of predictors (independent variables). Using this set of variables, the MLA generates a function using training data that maps inputs to desired outputs during training. The training process continues until the MLA achieves a desired level of accuracy on validation data. Unsupervised learning MLA does not involve predicting a target or outcome variable but learns patterns from untagged data. Such MLAs are capable of self-organization to capture patterns as probability densities, and are used e.g. for clustering a population of values into different groups. Clustering is used in many fields including pattern recognition, image analysis, bioinformatics, data compression, computer graphics, etc. Reinforcement learning MLA is trained to take actions or make decisions that maximize cumulative reward (e.g. a user-provided score). During training, the MLA is exposed to a training environment where it learns through trial and error to develop an optimal or near-optimal policy that maximizes reward. In doing so, the MLA learns from past experience and attempts to capture the best possible knowledge to make desirable decisions.

As will be appreciated by one skilled in the art, the present techniques may be embodied as a system, method or computer program product. Accordingly, the present techniques may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware.

Furthermore, the present techniques may take the form of a computer program product embodied in a computer readable medium having computer readable program code embodied thereon. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable medium may be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing.

Computer program code for carrying out operations of the present techniques may be written in any combination of one or more programming languages, including object-oriented programming languages and conventional procedural programming languages.

For example, program code for carrying out operations of the present techniques may comprise source, object or executable code in a conventional programming language (interpreted or compiled) such as C, or assembly code, code for setting up or controlling an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array), or code for a hardware description language such as VerilogTM or VHDL (Very high-speed integrated circuit Hardware Description Language).

The program code may execute entirely on the user's computer, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network. Code components may be embodied as procedures, methods or the like, and may comprise sub-components which may take the form of instructions or sequences of instructions at any of the levels of abstraction, from the direct machine instructions of a native instruction set to high-level compiled or interpreted language constructs.

It will also be clear to one of skill in the art that all or part of a logical method according to the preferred embodiments of the present techniques may suitably be embodied in a logic apparatus comprising logic elements to perform the steps of the method, and that such logic elements may comprise components such as logic gates in, for example a programmable logic array or application-specific integrated circuit. Such a logic arrangement may further be embodied in enabling elements for temporarily or permanently establishing logic structures in such an array or circuit using, for example, a virtual hardware descriptor language, which may be stored and transmitted using fixed or transmittable carrier media.

The examples and conditional language recited herein are intended to aid the reader in understanding the principles of the present technology and not to limit its scope to such specifically recited examples and conditions. It will be appreciated that those skilled in the art may devise various arrangements which, although not explicitly described or shown herein, nonetheless embody the principles of the present technology and are included within its scope as defined by the appended claims.

Furthermore, as an aid to understanding, the above description may describe relatively simplified implementations of the present technology. As persons skilled in the art would understand, various implementations of the present technology may be of a greater complexity.

In some cases, what are believed to be helpful examples of modifications to the present technology may also be set forth. This is done merely as an aid to understanding, and, again, not to limit the scope or set forth the bounds of the present technology. These modifications are not an exhaustive list, and a person skilled in the art may make other modifications while nonetheless remaining within the scope of the present technology. Further, where no examples of modifications have been set forth, it should not be interpreted that no modifications are possible and/or that what is described is the sole manner of implementing that element of the present technology.

Moreover, all statements herein reciting principles, aspects, and implementations of the technology, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof, whether they are currently known or developed in the future. Thus, for example, it will be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative circuitry embodying the principles of the present technology. Similarly, it will be appreciated that any flowcharts, flow diagrams, state transition diagrams, pseudo-code, and the like represent various processes which may be substantially represented in computer-readable media and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.

The functions of the various elements shown in the figures, including any functional block labeled as a "processor", may be provided through the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functions may be provided by a single dedicated processor, by a single shared processor, or by a plurality of individual processors, some of which may be shared. Moreover, explicit use of the term "processor" or "controller" should not be construed to refer exclusively to hardware capable of executing software, and may implicitly include, without limitation, digital signal processor (DSP) hardware, network processor, application specific integrated circuit (ASIC), field programmable gate array (FPGA), read-only memory (ROM) for storing software, random access memory (RAM), and non-volatile storage. Other hardware, conventional and/or custom, may also be included.

Software modules, or simply modules which are implied to be software, may be represented herein as any combination of flowchart elements or other elements indicating performance of process steps and/or textual description. Such modules may be executed by hardware that is expressly or implicitly shown.

It will be clear to one skilled in the art that many improvements and modifications can be made to the foregoing exemplary embodiments without departing from the scope of the present techniques.

Claims

1. A computer-implemented method of controlling a distributed energy provision system, the distributed energy provision system comprising a plurality of energy provision devices and a control module configured to control operation of the plurality of energy provision devices, the distributed energy provision system being arranged to deliver electrical power to a power grid, the method comprising : inputting first data into an optimisation algorithm to determine a power delivery parameter for each of one or more of the plurality of energy provision devices; sending a power delivery instruction to the one or more energy provision devices to operate the one or more energy provision devices for power delivery based on the respective power delivery parameter; for each of the one or more energy provision devices: obtaining second data from metering the energy provision devices during power delivery; inputting the second data into a device behaviour model of the energy provision device; estimating, using the device behaviour model, one or more device parameters of the energy provision device; and updating the optimisation algorithm using the one or more estimated device parameters respective of each of the one or more energy provision devices.

2. The method of claim 1, further comprising storing the second data in an observation database.

3. The method of claim 2, further comprising obtaining respective past second data for each of the one or more energy provision devices stored on the observation database and inputting the respective past second data into the respective device behaviour model for the one or more energy provision devices.

4. The method of any preceding claim, wherein the first data comprises manufacturer-specified device parameters for the one or more energy provision devices such as operation power, maximum/minimum state of charge, operation temperature, efficiency curve, response time, environmental parameters such as ambient temperature, humidity, or operational parameters such as operation costs, or any combination thereof.

5. The method of any preceding claim, wherein the second data comprises device parameters obtained through metering including operation power, operation temperature, state of charge, device configuration changes, or any combination thereof.

6. The method of any preceding claim, wherein the one or more device parameters of a respective energy provision device comprise an operation efficiency and/or a maximum operation power at a given state of charge.

7. The method of any preceding claim, wherein the device behaviour model of a respective energy provision device is configured to simulate operation of the respective energy provision device based on the second data obtained from the energy provision device, the meter associated with the energy provision device, and the power delivery instruction associated with the energy provision device.

8. The method of any preceding claim, wherein the power delivery instruction comprises an operation start time, an operation end time, an operation time period, an operation power at a given time, an average operation power, an overall energy to be delivered, or any combination thereof.

9. The method of any preceding claim, wherein the optimisation algorithm is a solution to a mixed-integer linear programming problem.

10. A computer-readable medium comprising machine-readable code, which, when executed by a processor, causes the processor to perform the method of any preceding claim.

11. A control module arranged to operate a plurality of energy provision devices for a distributed energy provision system to deliver electrical power to a power grid, the control module comprising one or more processors configured to execute machine-readable code stored on memory to perform the method of any of claims

12. A distributed energy provision system arranged to deliver electrical power to a power grid, comprising: a plurality of energy provision devices each configured to output electrical power; and a control module configured to control operation of the plurality of energy provision devices, the control module comprising memory configured to store machine-readable code and one or more processors configured to execute the machine-readable code, wherein execution of the machine-readable code causes the one or more processor to: input first data into an optimisation algorithm to determine a power delivery parameter for each of one or more of the plurality of energy provision devices; send a power delivery instruction to the one or more energy provision devices to operate the one or more energy provision devices for power delivery based on the respective power delivery parameter; for each of the one or more energy provision devices: obtain second data from metering the energy provision devices during power delivery; input the second data into a device behaviour model of the energy provision device; estimate, using the device behaviour model, one or more device parameters of the energy provision device; and update the optimisation algorithm using the one or more estimated device parameters respective of each of the one or more energy provision devices.

13. The distributed energy provision system of claim 12, wherein the distributed energy provision system is a virtual power plant.

14. The distributed energy provision system of claims 12 or 13, wherein the plurality of energy provision devices comprises one or more medium-scale power generating units, one or more micro combined heat and power units, one or more natural gas-fired reciprocating engines, one or more small-scale wind power plants, one or more photovoltaics units, one or more hydroelectricity plants, one or more biomass power plants, one or more backup generators, one or more energy storage systems and devices, or any combination thereof.