Method and system for controlling energy storage device or load

Abstract

A method (300) for controlling an energy storage device or load comprises generating a hybrid tariff based on a weighted combination of a tariff and network cost. An energy usage prediction and, where renewable or other generation sources are available, an energy generation prediction are generated (302,304) for time periods for each site in a plurality of sites. The method further comprises generating (306), for the site, a plan to selectively store energy in an energy storage device at the site, discharge energy from the energy storage device, or control when a load at the given site consumes energy based at least in part on optimising the cost of energy consumed according to the hybrid tariff based on the energy usage and generation prediction.

Description

[0001] METHOD AND SYSTEM FOR CONTROLLING ENERGY STORAGE DEVICE OR LOAD BACKGROUND OF THE INVENTION

[0002] Energy systems that include distributed energy resources such as solar panels, stationary battery storage, electric vehicle (EV) chargers, EV batteries, HVAC (heating, ventilation, and air conditioning), etc. are becoming increasingly prevalent in residential sites. The state of an installed energy system, such as the amount of solar being generated, when an EV may be plugged in, the amount of charge available in a battery storage, the amount of household consumption that is occurring, etc. is dependent on a variety of factors that themselves may vary. Thus, it can be challenging to determine how to best utilize self-generated energy and variable cost grid-provided energy within a home containing distributed energy resources in a way that matches the needs and goals of homeowners, as well as the utility power grid.

[0003] SUMMARY OF INVENTION

[0004] According to a first aspect of the present invention, there is provided a method for controlling an energy storage device or load, comprising:

[0005] receiving a tariff for a time period comprising plural tiers according to which the energy consumer at a site is charged for energy at different rates at different times; receiving a network cost profile for the time period;

[0006] generating a hybrid tariff for the time period based on a weighted combination of the tariff and network cost;

[0007] generating an energy usage prediction and, where renewable or other generation sources are available, an energy generation prediction for the time period for each site in a plurality of sites, wherein the plurality of energy generation scenarios have different energy generation scenario likelihoods;

[0008] generating, for the site, a plan to selectively store energy in an energy storage device at the site, discharge energy from the energy storage device, or control when a load at the given site consumes energy based at least in part on optimising the cost of energy consumed according to the hybrid tariff based on the energy usage and generation prediction.

[0009] As discussed below, this allows the system to develop an energy storage system (e.g. domestic storage battery or EV) or load control plan, i.e. when to charge, discharge, consume energy etc for an upcoming time period that co-optimises for at least two factors, namely the price the end user pays for the energy according to their tariff, i.e. contract with the utility company, and the network cost, i.e. a price signal or other cost or allowance the utility company, generator or another grid actor incurs for the energy used. This leads to cost saving for the utility and more efficient use of resources. In particular, the total value delivered is preferably greater than the value delivered by optimising only for one or the other. As shown below, this can deliver cost savings to the utility of approx. 4% and keep costs for the customer neutral or even reduce costs at the same time. Alternatively, the utility costs could be reduced by approx. 7% to 8% whilst only reducing the customer's savings by approx. 1%.

[0010] The use of residential energy storage systems / devices for demand response services is an effective way of providing various services to the grid, e.g. frequency response, firming renewable supplies, etc. Various signals may be provided to residential sites which may act as proxies for control signals to drive certain behaviours at the end sites, either user behaviour or automatic response by devices at the end sites, e.g. using electricity at off peak times. For instance in prior art systems an end device such as an energy storage system may optimise its behaviour based on the consumer tariff (so called "Behind the Meter" optimisation) so as to minimise cost to the consumer.

[0011] However, optimising based purely on consumer cost may not necessarily be optimum in terms of the price the utility pays wholesale for the energy needed to supply the customer, and hence the control plan is inefficient in making use of the energy. Similarly, controlling a device for the benefit of the utility company based on a wholesale price or other network cost whilst ignoring the end user tariff may lead to savings for the utility but increased costs or at least unacceptable increased costs for the end user. The invention creates a hybrid tariff which includes a weighted component related to the network cost for use in the optimisation scheme such that greater cost and energy use efficiencies are achieved without having to completely redesign the optimisation scheme.

[0012] Furthermore, the invention makes co-optimisation seamless, removing the need for explicit DR events. Instead the invention co-optimises implicitly as part of regular BTM operation, i.e. using the hybrid tariff in place of the regular tariff, and in the process reduces the secondary cost function with only minimal effect on the primary customer cost. Thus, there is a joint financial benefit but also efficiency of operation.

[0013] Another advantage is that by co-optimising via constructing a hybrid tariff for otherwise standard BTM optimisation, it means that the optimisation algorithm employed by the optimisation engine can be changed without it affecting the hybrid optimisation. For instance, the algorithm could be changed from being based on linear algebra to genetic algorithms based on the hybrid tariff, whereas with a constraint-based two-step optimisation the approach would need to be fundamentally different requiring the optimisation engine to be completely overhauled.

[0014] The energy storage system may for instance be a domestic rechargeable storage battery possibly used optionally together with solar or other renewables at the end site. An Electric Vehicle battery connected to a charging station at the end site could also be used. In some examples, the controlled device is a load where its consumption can be controlled e.g. by time shifting when the load is activated. In cases where there is no generation on site, i.e. no renewable sources, then the generation prediction may be set to zero in the simulation.

[0015] In embodiments, the network cost is based on one or more of:

[0016] wholesale price;

[0017] carbon price; and

[0018] imbalance settlement price.

[0019] The wholesale price is the market price paid by the utility company for the power. The carbon price reflects the cost of generating power based on fossil fuels. The Imbalance Price is used to settle energy imbalance volumes.

[0020] In principle any price or cost signal used in the energy network could be used as the network cost, i.e. secondary factor, in the hybrid tariff with the customer tariff preferably being the primary factor. This is typically a price the utility company pays in providing the energy either at a grid level or local grid level where prices are different in different parts of the grid.

[0021] In an embodiment the weighted combination is according to the formula:

[0022] hybrid tariff = (1-x) * consumer tariff + x * network cost, where the weighting factor x is in the range 0 < x <1.

[0023] In an embodiment x is between 0.05 and 0.5. In an embodiment x is between 0.1 and 0.25. These values are found to work well where the network cost is the wholesale price or some other tangible monetary value associated with the cost of supplying power. It will be appreciated that different values of x or weighting formulae may be preferred for other use cases and that the weighting factor may be determined empirically or via simulation as described below in this disclosure.

[0024] In general, these values meet the objective is that the customer costs are the primary objective and minimising for the network costs is a secondary objective. Thus, the network cost provides relatively small perturbations to the primary cost which is the customer tariff.

[0025] In other embodiments, hybrid tariff = consumer tariff + x * network cost, where 0 < x <1. In an example, the network cost is the carbon price which is based on a combination of Carbon Price Floor * Grid Carbon Intensity. In effect the Carbon Price Floor is the weighting factor in this example. Carbon Price Floor is a minimum carbon price imposed by the government, e.g. in the UK, which aims to prevent the price paid for carbon emissions falling below a minimum price. Grid carbon intensity is a measure of how much carbon dioxide (CO2) is released to produce each kilowatt hour (kWh) of electricity which changes throughout the day, season, and by hour due to changes in electricity demand and the amount of low carbon generation and so can provide perturbations to the primary factor, i.e. customer tariff.

[0026] In an embodiment, the method comprising simulating energy usage at at least one end site across multiple days of historical data and multiple weighting values for the weighted combination to obtain an optimum value of the weighting values. Simulations may be run on one or more end sites to ascertain an optimum value of the weighting values for the site or location or time period, e.g. taking account of factors such as size of battery and solar panels, expected generation and / or consumption taking into account expected weather conditions, etc. Such simulations may be periodically updated or tailored to specific sites.

[0027] In an embodiment, the optimum value of the weighting values minimises cost to the utility without increasing cost to the consumer. Typically a value of x of approx. 0.2 +- 25% allows the battery to optimally charge during low cost periods during the consumer tariff and so avoid increasing cost to the consumer, whilst minimizing cost, e.g. wholesale cost, within this constraint.

[0028] In an embodiment, optimum value of the weighting value minimises cost to the utility whilst increasing cost to the customer by no more than 5%. In an embodiment, the method comprising calculating the cost saving to the utility company and the cost difference to the consumer. Thus, the cost to the utility company may be further reduced and the loss to the consumer quantified such that they can be compensated or otherwise persuaded to participate. As described below, the opportunity cost of co-optimisation compared with behind the meter optimisation can be used to select which batteries participate in the scheme, i.e. using those with a lower opportunity cost.

[0029] In an embodiment, the hybrid price during any particular tariff tier is never more expensive than in a lower tariff tier and vice versa. In other words, the weighting is determined, e.g. empirically or by simulation based on historical data, such that the secondary factor, i.e. network cost, applies relatively small perturbations to the primary factor, the customer tariff, such that the price bands for the respective tiers in the hybrid tariff remain distinct and non-overlapping. Thus, for instance, if the smallest difference between price bands is Xmin and then the secondary factor may be weighted such that the expected maximum variation based on historical and / or current data remains below Xmin, or some proportion of it, e.g. between 10% and 25% of Xmin.

[0030] In an embodiment, the network cost is normalised against the customer tariff before applying the weighting.

[0031] In an embodiment, the method is carried out separately for different end sites.

[0032] In an embodiment, the hybrid tariff is calculated separately for each energy consumer. In an embodiment, the network cost profile is localised such that different network cost profiles are used for different consumers in different parts of the network.

[0033] In an embodiment, the predictions and optimisation is performed separately for different end sites.

[0034] According to a second aspect of the present invention, there is provided a system, comprising:

[0035] a processor configured to:

[0036] receive a tariff for a time period comprising plural tiers according to which the energy consumer at a site is charged for energy at different rates at different times;

[0037] receive a network cost profile for the time period;

[0038] generate a hybrid tariff based on a weighted combination of the tariff and network cost; generate an energy usage prediction and, where renewable or other generation sources is available, an energy generation prediction for the time period for each site in a plurality of sites;

[0039] generate, for the site, a plan to selectively store energy in an energy storage device at the site, discharge energy from the energy storage device, or control when a load at the given site consumes energy based at least in part on optimising the cost of energy consumed according to the hybrid tariff based on the energy usage and generation prediction.

[0040] In an embodiment, the system comprising a communications interface arranged to dispatch a charging plan to a remote energy storage device or load.

[0041] In an embodiment, the system comprising plural remote energy storage devices and / or loads at plural end sites in communication with the processor via a network. The system may thus control a fleet of storage devices and loads across a local or wider area.

[0042] The predictions may include for each site a plurality of energy usage scenarios having different energy generation scenario likelihoods and a plurality energy generation scenarios having different energy generation scenario likelihoods wherein the plural scenarios are used in the optimisation to generate the plan. For instance, as described herein, a random walk may be performed to generate plural scenarios and the optimisation engine may develop plural candidate battery control plans for the respective scenarios from which a final control plan may be selected for dispatch to the storage device / load.

[0043] Another aspect relates to a method for controlling an energy storage device or load, comprising:

[0044] receiving a tariff for a time period representing the cost of energy at different times;

[0045] generating an energy usage prediction and, where renewable or other generation sources are available, an energy generation prediction for the time period for each site in a plurality of sites,

[0046] generating, for the site, a plan to selectively store energy in an energy storage device at the site, discharge energy from the energy storage device, or control when a load at the given site consumes energy based at least in part on optimising the cost of energy consumed according to the tariff based on the energy usage and generation prediction and weighted according to efficiency of consuming the energy.

[0047] In another aspect, a system for carrying out the method is provided.

[0048] In an embodiment, plural energy storage devices or loads are controlled, such as a storage battery and an EV, and the efficiencies include the efficiencies of electrical conversions required in providing solar energy to those devices or loads.

[0049] BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Various embodiments of the invention are disclosed in the following detailed description and the accompanying drawings.

[0051] FIG. 1 A illustrates an embodiment of an environment in which predictive energy provisioning is performed.

[0052] FIG. 1 B illustrates an embodiment of a home energy system.

[0053] FIG. 2A illustrates an embodiment of forecast traces.

[0054] FIG. 2B illustrates an embodiment of a battery storage state of charge according to a battery control plan.

[0055] FIG. 2C illustrates an embodiment of an intent.

[0056] FIG. 2D illustrates an embodiment of an intent.

[0057] FIG. 2E illustrates an embodiment of an intent.

[0058] FIG. 3 illustrates an embodiment of a process for generating an energy storage device control plan.

[0059] FIG. 4 illustrates an embodiment of a process for executing an energy storage device control plan.

[0060] FIG. 5 is a flow diagram illustrating an embodiment of a process for co-optimization. FIGs. 6 to 10 illustrate the effects of a hybrid tariff optimisation method on various parameters of a home energy storage solution; and

[0061] Fig 11 shows a system for carrying out the method of Figs 6 to 10.

[0062] DETAILED DESCRIPTION

[0063] The invention can be implemented in numerous ways, including as a process; an apparatus; a system; a composition of matter; a computer program product embodied on a computer readable storage medium; and / or a processor, such as a processor configured to execute instructions stored on and / or provided by a memory coupled to the processor. In this specification, these implementations, or any other form that the invention may take, may be referred to as techniques. In general, the order of the steps of disclosed processes may be altered within the scope of the invention. Unless stated otherwise, a component such as a processor or a memory described as being configured to perform a task may be implemented as a general component that is temporarily configured to perform the task at a given time or a specific component that is manufactured to perform the task. As used herein, the term ‘processor’ refers to one or more devices, circuits, and / or processing cores configured to process data, such as computer program instructions.

[0064] A detailed description of one or more embodiments of the invention is provided below along with accompanying figures that illustrate the principles of the invention. The invention is described in connection with such embodiments, but the invention is not limited to any embodiment.

[0065] The scope of the invention is limited only by the claims and the invention encompasses numerous alternatives, modifications and equivalents. Numerous specific details are set forth in the following description in order to provide a thorough understanding of the invention. These details are provided for the purpose of example and the invention may be practiced according to the claims without some or all of these specific details. For the purpose of clarity, technical material that is known in the technical fields related to the invention has not been described in detail so that the invention is not unnecessarily obscured.

[0066] Described herein are embodiments of predictive energy provisioning. In various embodiments, predictive energy provisioning includes generating, on a per-site basis, predictions of solar power generation, predictions of energy consumption (including, for example, predictions of when an EV may be plugged in or out). Using such predictions, control of distributed energy resources (such as a battery, EV charger, or HVAC) at the site is facilitated (e.g., parameters for charging / discharging the device, or shifting its energy usage temporally are determined). Further, in various embodiments, the predictions are used to power or support interfaces that a user may interact with to understand the state of their home energy system, where, in addition to being provided insight into historical information about their site’s power system, they are also able to View forecasted information about the future behaviour or state of their home energy system.

[0067] While embodiments involving predictive power provisioning at residential sites (e.g., households) are described herein for illustrative purposes, the predictive provisioning techniques described herein may be variously adapted to any other type of site, as appropriate.

[0068] FIG. 1 A illustrates an embodiment of an environment in which predictive energy provisioning is performed.

[0069] In this example, sites 102 and 104 are residential sites that include home energy systems. FIG. 1 B illustrates an embodiment of a home energy system. In this example, the home energy system includes a local power generation system (e.g., an array of photovoltaic PV) panels 152) and an energy storage device 154. In this example, the energy storage device includes batteries. The techniques described herein may be variously adapted to accommodate any other type of energy storage (e.g., fuel cells), as appropriate. Each site also has a home energy management system 156 that manages the distribution of power from the PV array 152, battery storage 154, and the grid 158 to meet the power demands of the home. In this example, the home energy management system also controls energy usage of controllable loads of the home, such as electric vehicles (e.g., EV 160) and HVAC units (e.g., air conditioner 162, heaters, etc). Such control includes, for example, temporally shifting energy usage of such loads, or otherwise controlling when a particular device consumes energy. For example, in some embodiments, the loads are connected to controllable circuits or switches that are under the management of the home energy management system. In some embodiments, the home energy management system is configured to selectively control power delivered to such loads by opening or closing or otherwise modulating the switches (which are, for example, smart switches that are under the control of the home energy management system). This effectively results in control of when the loads / devices are connected to, or disconnected from, power (and whether they are drawing power or consuming energy). In other embodiments, the loads include smart devices that are in communication with the energy management system (e.g., over wireless protocols such as WiFi), where such smart devices are directly under the control of the energy management system (where they may be directly commanded to turn on or off). As shown in the example of FIG. 1A, energy provisioning planning system 106 is configured to determine, for each site, site-specific forecasts of solar power generation and energy consumption. The energy provisioning planning system is also configured to, using the forecasted predictions of solar power generation and energy consumption, determine an energy storage device control plan for optimally controlling the battery storage (e.g., controlling when and how the battery storage is cycled) at a residential site. In some embodiments, energy provisioning planning system 106 communicates with each of the home energy systems at the various sites via a network such as the Internet 108. In this example, the customers at each of the residential sites (e.g., homeowners of the sites) are able to View information regarding their home energy systems, such as via mobile applications installed on their phones (e.g., mobile devices 110 and 112), which are in communication with the energy provisioning planning system. Customers or users are also able to view the information about their home energy systems via browserbased web interfaces (e. g., when using devices such as laptop 114) using information provided by the energy provisioning planning system.

[0070] In one embodiment, the energy provisioning planning system is implemented on a scalable platform to allow determination of forecasts and device control plans using operations that are performed on data from a large volume of devices (e.g., site information such as grid meter information, solar panel power measurements, historical battery storage system information, etc.) and data sources (e.g., data sources 116, such as weather service information). As one example, the energy provisioning planning system is implemented on a cloud architecture such as Amazon AWS or Google Cloud. In this example, energy provisioning planning system 106 includes energy consumption prediction engine 118 and solar generation prediction engine 120, which are configured to, respectively, generate machine learning predictions, on a per-site (e.g., per residential site or household) basis, of that site’s projected energy consumption and power generation (e.g., solar generation) for a given time period (e.g., given day). As will be described in further detail below, the generation of the energy consumption and solar production\predictions is based on ingestion and analysis of various types of information. In various embodiments, machine learning techniques are used to generate the predictions described herein. For example, supervised machine learning regressions algorithms such as support vector regression (SVR) are used.

[0071] In some embodiments, the output of the machine learning models is not a specific number of kilowatts of solar production or energy consumption at a particular future time. Rather, the output is statistical, such as a probability distribution that reflects the stochastic and real -world nature of the data used to make the predictions. For example, the probabilistic output is used to account for the variability in energy consumption. For example, while energy consumption predictions may be based on a customer's past behaviour, they may deviate, for example, by having a large party that was unpredicted. As will be described in further detail below, the predictive models described herein are built taking into account real-world uncertainty. This includes accounting for over-fitting, further details of which will be described in further detail below.

[0072] As one example, the probability distribution is reflected in three lines, with different options, such as a min-value prediction, a max prediction, and a median prediction. Those predictions are used as input for smart device planning, which includes predictions, as well as the customer’s rate plan, in order to minimize their cost or their carbon footprint, or a mixture of both.

[0073] The energy provisioning planning system further includes device control plan optimization engine 122, which is configured to, based on the aforementioned predictions, determine, for each household, a plan for controlling the usage of the household’s energy storage device (e.g., when to charge and / or discharge the battery storage, for what amount of time, at what power level, etc.). The device control plan may be optimized for various scenarios. For example, device control plan optimization engine 122 further includes site-level optimization engine 124 and fleet level optimization engine 126. The site-level optimization engine 124 is configured to determine a device control plan that is optimal for a specific site, independent of other sites.

[0074] Fleet level optimization engine 126 is configured to determine an optimal device control plan in the context of a fleet level dispatch of multiple sites, such as to provide a Virtual power plant. In this example, on a periodic basis (e.g., each day), the energy provisioning planning system is configured to generate, for each site, the aforementioned solar power generation forecasts, energy consumption forecasts, and device control plans.

[0075] In some embodiments, the solar generation and site consumption predictions, as well as device control plans, are generated for intervals of time, such as 30 minute predictions (predictions for different windows or granularities of time may also be made, as appropriate). In some embodiments, projections are made for a future period of time, such as the next 48 hours. For example, when projecting 48 hours out into the future, 96 half-hour predictions are made. In some embodiments, a prediction run is made each day of forecasts for the next two days. Other intervals of running the predictions may be used. Other time periods for each prediction run may also be used.

[0076] In some embodiments, the predictions are rerun when triggered by an event. For example, if a deviation from a prediction exceeds a threshold, then the prediction and planning is rerun. For example, if the energy consumption profile deviates by a threshold amount from the expected energy consumption planner, the planning and modeling of predictions is rerun.

[0077] Further details regarding prediction of solar generation and prediction of energy consumption, as well as device planning and control, are described in further detail below.

[0078] Forecasting Solar Power Generation

[0079] In some embodiments, the solar generation prediction engine 120 is configured to connect with various data sources, ingest data from the connected sources, and normalize such ingested data into a usable format. The formatted data is then evaluated to determine a forecast of the expected solar power generation over a time period such as the next 24 to 48 hours.

[0080] In some embodiments, the solar generation prediction engine executes a machine learning algorithm that generates a function or equation that, given an input set of variables, provides as output an amount of solar power / energy generated for a time frame (e.g., a day). The machine learning model is trained on historical information collected on a per-site basis. For example, the amount of solar power generated by the site’s PV array is metered and measured over time. This results in the collection of historical information about solar production over time. In addition to the collection of solar production information, metadata corresponding to the time at which solar production measurements were taken is also recorded. Such metadata includes weather information such as UV radiation. In some embodiments, the metadata that is used in the training of the solar production forecast algorithm corresponds to the input parameters that will be used when requesting a solar production prediction to be made. In this way, historical solar production information and corresponding contextual metadata are used as training data to train the machine learning model, wherein the trained machine learning model is used to predict an amount of solar generation at a site for a period of time.

[0081] The following are example factors for determining solar power generation:

[0082] solar irradiance, humidity, site orientation, array size, peak power generation of the solar array (e.g., kilowatt peak), coordinates (e.g., latitude / longitude) of the site, or other location data of the site, historical measures of power generated by the solar panel array at the site. For example, data collected from metering of the power generated by a PV array is used to generate an inference model for predicting PV power generation.

[0083] The prediction of solar generation (and its accuracy) is a function of information regarding the home site.

[0084] As described above, such information used to perform forecasting includes weather forecasting information, where feeds from weather forecasting engines are used as input (e.g., to obtain information such as solar irradiance, solar flux, humidity, cloud cover, etc.). Historic information is also used, as described above, such as any site-specific information learned from observation of the specific site.

[0085] For example, solar generation of a system is monitored over time. Patterns in the behaviour of the solar generation are determined from the historical solar generation data. One example of a pattern that can be derived from historical solar production data is that typically, the home has a low amount of solar generation at 4 PM because it is shaded (where this may not be true for another home situated similarly).

[0086] In some embodiments, to improve the accuracy of the solar generation prediction model, the predictions are continuously compared against measured reality (e.g., the training data is updated with new data as it is collected).

[0087] In some embodiments, solar generation predictions are generated even if there is not any site-specific power data. For example, suppose that solar metering is not available for a site of interest (and thus historical solar production is unable to be evaluated). In some embodiments, factors such as site orientation (which may be determined from publicly accessible satellite imagery) are used to generate a solar production forecast. As another example, the solar metering data for another similar site is used as a factor in predicting solar generation for the site of interest.

[0088] As another example, suppose that there is no metering on solar panels, but EV charging is to be matched to when there is solar power available. In some embodiments, based on the location of the house, the size of the solar array, the cardinal orientation of the panels, weather forecasts (e.g., ultraviolet radiation forecasts), etc., a curve is generated of the solar power generated by the panels. Matching may then be performed using public or other sources of data.

[0089] Such a prediction may also be used to provide a detailed solar generation forecast to a customer before they have even purchased or installed any hardware. For example, based on the geographic region of the home, and a given array size, a day-by-day prediction or trace of solar power generation for a given location is provided.

[0090] In some embodiments, the forecast is statistical in nature. For example, rather than predicting a specific amount of power generated by the solar panel array at a point in time, a probability spread or distribution of the amount of solar power production is generated for a future window of time. Statistical measures with respect to the forecast are also determined, such as the 10th, 90th, and median (50th) percentile solar power generated. In some embodiments, the difference between the 10th and 90th percentile solar production predictions is determined as a measure of variability or risk in the solar production forecast. For example, the tighter or smaller the spread between the 10th and 90th percentile predictions, the higher the confidence in the prediction. The wider the spread of the 10th and 90th percentile predictions, the lower the confidence in the prediction. As device control plans will be determined based on the forecasted solar production, the variability measures are used as a measure of reliability when generating device control plans. As will be described in further detail below, creation of an optimal device control plan also takes into account the amount of variability in the underlying predictions upon which a device control plan relies. This similarly applies for energy consumption forecasts, as will be described in further detail below.

[0091] FIG. 2A illustrates an embodiment of forecast traces. Examples of projected solar production traces overtime (which show, for example, 10th, median, and 90th percentile PV power production over a 48-hour period) are shown at 202 of FIG. 2A (above the 0 W line), where the 10th percentile corresponds to the production-min forecast, the 90th percentile corresponds to the production-max forecast, and the median corresponds to the production forecast.

[0092] As shown in this example, over the course of the future 48-hour period, the solar production traces spread out, come together, then spread out again. Consider the time period shown around midnight (00:00) on 8 / 17. During this time period, there is a larger spread in the solar production traces. This reflects the probability distribution of the predicted solar production forecast, where due to uncertainty or variability, the range in which the true solar generation is likely to fall is larger (and there is less certainty that reality will match the prediction in that time period, so there is a broader range for the prediction). This is in contrast to the period between 12:00-18:00 on 8 / 17, where in the middle of the night, there is high confidence that there will not be any sun, and no solar generation, and so the actual production is likely to fall within a narrower range of potential production values (smaller difference between the 10th and 90th percentile predictions). After the nighttime period, the traces spread back out again during the day, reflecting the variability that may occur in solar production (e.g., due to various random outliers). For example, during the day, cloud cover, the UV or solar radiance, etc. may fluctuate. As one example, what may occur in reality is that the weather is sunny at one period, in which the solar panels are generating a certain amount of power, and then clouds roll across the sky for 15 minutes, causing a dip in solar generation, where the generation then increases again when the clouds pass. Thus, actual production is likely to fall in a wider range of values.

[0093] In some embodiments, to capture these fluctuations that may occur at various parts of the day, predictions are made at a half hour interval. Predictions at larger or smaller intervals of time may also be made.

[0094] Forecasting Energy Consumption

[0095] In some embodiments, energy consumption prediction engine (118) is configured to generate a forecast of an amount of energy that will be consumed by a site over a period of time.

[0096] The following is an example of generating a prediction of energy consumption for a site. In some embodiments, energy consumption data for a historical or previous period of time for a site is obtained. For example, the last two weeks of data for that site is obtained (other periods of data may be utilized). Historical data over all time is not necessarily evaluated, as there may be changes in patterns due to seasons (e.g., historical winter data may not be beneficial in making a prediction for summer energy usage).

[0097] Suppose that a prediction is to be made for a given day, for example Tuesday. In this example, a View of the consumption of the past two weeks is obtained and used to build a consumption model, including a bias for recent days. As well as this, patterns of differences in energy usage for different days are also determined (e.g., users may tend to use more energy on Friday than other days of the week, or have different patterns due to different schedules).

[0098] In some embodiments, weather patterns are also evaluated. For example, the current weather is evaluated and used to bias what historical data is utilized. Suppose that it is currently rainy and cold. Historical data for the last 14 days is obtained. It is determined what type of weather was experienced on the last 14 days. A comparison of similarity is determined between the weather on those days and the current weather. Suppose that during the two week period it was sunny for 10 days, with the last four days being rainy and cold, similar to today. In this case, rather than averaging the consumption and generation historical data across the last 14 days, the historical data from the last 4 days, where the weather is more similar to today’s, is used to generate today’s forecast.

[0099] Relationships between weather patterns and consumption patterns are also established (e.g., observing that air conditioning loads are higher when the weather is hotter).

[0100] Such contextual weather information is also used to determine consumption productions. Consumption predictions may also be adjusted based on other contextual information. For example, if a customer indicates that they will be on vacation (e.g., by selecting for the system to enter a “vacation” mode), then the consumption prediction for the corresponding time period is reduced.

[0101] The amount of historical data used (e.g., the time period of interest) may be tuned to account for variability such as seasonality. In some embodiments, the historical data is used to predict the likely medians and outliers for future consumption. In some embodiments, a weather reference is added to the consumption prediction / forecasting. For example, if it is known that tomorrow (for which a prediction is being made) will be unseasonably hot, then it is anticipated that there will be an air conditioning load that will increase the consumption tomorrow. That is, when generating a consumption prediction, in addition to historical consumption data, weather information for the day being predicted is used to determine whether to augment the historical consumption data with additional or fewer loads (more or less consumption) based on the weather forecast. In some embodiments, historical consumption data is determined in various temporal granularities. For example, while predictions for thirty-minute increments or intervals of time are made, the historical consumption data may be gathered at smaller intervals, such as every 30 seconds, every minute, every 5 minutes, etc. In some embodiments, the historical consumption is a derived value that is derived from the historical behaviour of the home energy system, such as the battery state of charge, PV power production, and grid import / export.

[0102] Examples of historical data that are collected and used for forecasting future consumption include:.

[0103] The amount of battery charging / discharging measured in kilowatts

[0104] The amount of solar power generated in kilowatts. For example, for a site, metering of the power generated from the solar panels is used to generate an inference model The amount of power delivered by the inverter

[0105] The amount of power imported from or exported to the grid

[0106] The above historical home energy system behaviour is collected in various ways, such as via metering of energy sources such as the PV array, battery storage, and grid. As described above, the various measurements that are collected may be at various levels of temporal granularity (e.g., reported at one second, one minute, etc.).

[0107] From such historical measurements of power distribution, the amount of energy consumed for a given period of time is derived. For example, if there is 4 kWh of solar being generated, and 2 kWh of consumption, then the 2 kWh of excess solar will either go to charging the battery or be delivered to the grid so that the source power and sinked power net to zero.

[0108] In some embodiments, energy consumption predictions at a household / site level are determined. In some embodiments, metering of specific devices or specific circuits (e. g., kitchen, bedroom, etc.) is performed, in which the predictions may be made at those granularities or segments.

[0109] In some embodiments, machine learning predictions of when users connect or disconnect (e.g., plug in / out) their EVs are also performed. This improves scheduling of how that EV should be charged in a home site with PV and / or ESS (energy storage systems such as battery storage).

[0110] As one example, historical EV connection and disconnection events (e.g., collected by metering the EV) are aggregated and used to construct a statistical measure (e.g., probability density function GDP» of the probability of disconnecting the EV at a certain time after a duration of being connected. By doing so, predictions may be made when an EV is to be charged, at what rate, and for how long. As the price of electricity may vary, being able to predict when an EV is charged provides another input to the system to determine various approaches to how to optimize energy stored in a battery. For example, if the predicted EV plugin time were late at night during a cheaper energy period, then a different approach to utilizing stored battery charge would be used then if the predicted plugin time were during an expensive period of grid energy. In this way, when an EV is predicted to be charged informs how energy will be consumed, and is used to optimize (e.g., minimize) the amount of usage from the grid at expensive periods of time.

[0111] In some embodiments, generating a prediction of energy consumption includes generating power consumption curves. In some embodiments, a power curve indicates an amount of power consumed as a function of time. For example, the power curve indicates the forecasted amount of power consumed as a function of time over a period of 48 hours. The power curve may be queried at a certain temporal granularity. For example, average power consumed for half hour blocks of time are determined. This results in 96 power consumption values over a 48-hour period.

[0112] The historical data may be accumulated or aggregated for half hour windows of time (e.g., to determine the average power draw over a half hour period, the average battery state of charge, the average solar generation, etc.).

[0113] The averaging allows for a smoothing of the data over a period of time (where the actual trace of electricity will be jagged and have spikes every time a device is turned on or off). In some embodiments, predicting energy consumption at a site includes performing risk profiling to account for variability in energy usage at a site. For example, users of a household may be variable in their energy usage, or use energy in a varied way. For example, a bank manager may have a consistent schedule in when they are home and using energy, while this may not be the case for a freelance journalist who may be in and out of the home in a more arbitrary manner given the nature of their assignments.

[0114] This variability (which is site-to-site) influences the manner in which the system generates device control plans (as the device control plans rely on the energy consumption forecasts, similarly to the solar production forecasts). In some embodiments, the variability of the energy consumption is determined using statistical metadata pertaining to the energy consumption forecast. For example, the energy consumption forecast is generated as a probability distribution. In some embodiments, 10th, 50th (median), and 90th percentile energy consumption values are determined. As one example, a risk profile measuring the variability of the energy consumption is determined as the difference between the 10th and 90th percentile energy consumption predictions.

[0115] For example, a median (50th percentile) predicted energy consumption is generated. That is, there is a 50% probability that the actual consumed energy will be less than the median value, and there is a 50% probability that the actual consumed energy will be greater than the median value. As another example, 10th and 90th percentile energy consumption predictions are generated, where there is a 10% likelihood or probability that the amount of energy consumption will be below the 10th percentile value, and a 90% probability that the amount of energy consumption will be below the 90th percentile value. A larger gap between the 10th and 90th percentile values indicates larger variability in the historical energy consumption data (i.e., that the actual energy consumption will likely be in a broader range of values, and is unlikely to be in a narrow band of values).

[0116] The probability distribution that is generated as output indicates, for a particular time period, the likelihood of having predicted the correct amount of energy consumption within that period. This includes the likelihood that the 10th percentile line is lower than reality, or the likelihood that the actual energy consumption is below the 90% line (90th percentile value).

[0117] Examples of projected energy consumption traces over time (which show, for example, 10th, median, and 90th percentile PV power production over a 48 hour period) are shown at 204 of FIG. 2A (below the OW line), where the 10th percentile corresponds to the consumption-min forecast, the 90th percentile corresponds to the consumption-max forecast, and the median corresponds to the consumption forecast.

[0118] As illustrated in this example, there is a large amount of variation in the amount of power that is consumed at the site. For example, for the same period of 7 PM-9 PM on the two different days covered by this 48-hour plan, the consumption of power will be different (e.g., different between Thursday and Friday, where on Friday people may go out to eat rather than being at home for dinner on Thursday). Other periods may be relatively the same across different days, such as in the middle of the night. As shown in this example, average or aggregate power predictions are made for half hour time buckets.

[0119] Device Control Plan Generation The solar generation and energy consumption predictions / forecasts described above are used to determine plans for controlling usage of the battery at a site. For example, an optimal plan for controlling a site’s battery storage is generated based on the predicted solar generation and energy consumption, as well as any economic factors (e.g., tariffs, net metering, time of use rate plans, etc.), and any desired goals (e.g., grid usage minimization, economic benefit maximization, etc.). Other examples of goals include behind-the — meter optimization for wholesale market pricing so that asset owners can manage their wholesale procurement costs. For example, the variable pricing faced by an energy utility in the wholesale market is evaluated. For example, one model is that a utility leases a battery to their customer. The customer has a flat rate plan, but the utility has the ability to charge and discharge the battery as desired.

[0120] For example, the homeowners may be on plans in which energy is provided as a service, where rather than paying based on the amount of energy consumed from the utility, they pay for access to a certain capacity. While the charging / discharging of the battery would not affect the end customer bill as they are on a flat rate, the utility company can utilize the energy in end customer’s homes in a manner that lowers how much, and when they procure energy from a wholesale market (which may also have variable energy pricing). For example, if the batteries have been charged up off of solar power, then at night, when the wholesale market for energy is expensive, the grid utility can discharge those batteries to supply demand from homeowners, rather than having to procure energy from the wholesale market to meet demand.

[0121] In this way, the utility does not need to procure as much power on the open market. This results in a cost savings to the utility, and allows the utility to manage the cost of procuring energy. As another example, the battery devices may be controlled to charge or discharge in a manner so that the utility grid can manage the type of energy to be procured from the wholesale market, so as to reduce carbon emissions. For example, the grid can discharge home batteries to meet energy demand when the energy from the wholesale market is provided using more carbon intensive sources (e. g., coal). In this way, the utility grid can reduce the carbon intensity of the energy it is procuring. The utility grid may then prioritize delivering grid power to homes when the energy the utility is procuring from the wholesale market can be delivered via renewable resources (e.g., from solar farms, wind farms, etc.).

[0122] As another example, asset health / degradation cost (e.g., impact on life of battery storage) can be taken into account during optimization. For example, a system health cost is assigned when making charging decisions, in order to take into account the long term health effect of a charging decisions. For example, higher charging (or discharging) rates may have a greater impact on degrading or lowering overall capacity of a battery device. In this way, the state of health cost is taken into account when generating a device control plan.

[0123] As shown in the above, the optimization process for determining device control is flexible, allowing optimization to be performed according to various different types of objective functions.

[0124] In some embodiments, a plan is a control plan for controlling the behaviour of the battery. This includes a plan for specifying when the battery is to be idle, when the battery is to be charging from solar, when the battery should be charging from low cost energy, when the battery should be discharging energy into home loads, the time at which battery charging / discharging should be performed taking into account energy tariffs, the different costs of energy at various times of day, the differential cost of solar, potential benefit from exporting energy to the grid, etc.

[0125] For example, suppose that a goal is to maximize cost saving during a future period of time for which grid energy will be expensive. Suppose that it is predicted that there will be 7 kWh of consumption in that future period of time. A control plan is generated to ensure that there is at least 7 kWh of energy stored in the battery by the time of that price peak. To do so, a set of actions is taken to facilitate that amount of energy being available in the battery, at that time.

[0126] The battery’s state of charge is the result of actions that are taken and the way that the battery is controlled, rather than events that will simply occur. The state of charge is an output of a View of the future given how the battery device is controlled to achieve various targets or goals.

[0127] Using the predictions generated above, plans for controlling usage of the battery (e.g., to charge or discharge the battery) are generated. For example, when the predictions of generation and consumption are generated, they are treated as a reality of the future. For example, a plan for when to charge the battery and by how much over time is determined based on when there is expected to be solar generation, when consumption is to occur, etc. In some embodiments, the solar production and consumption forecasts are run against another algorithm which generates numerous potential series of actions that a home site could take (e.g., charging and / or discharging of batteries at various times for certain durations at certain power levels), and determines an optimal delivered value. For example, the planning algorithm generates a plurality of candidate battery control plans. For each battery control plan, an expected value generated if the plan were executed is determined.

[0128] In some embodiments, the device control plan optimization engine is implemented using a linear solver that has an objective function that is to optimize against a goal, such as reducing cost and / or maximizing benefit. The linear solver is provided various inputs, such as the size / capacity of the battery storage on the site (e.g., the kWh the battery can hold), the maximum kilowatt discharge rate of the inverter of that battery, the round trip efficiency of the battery (e.g., if charging up by 100 kW, how much of that can be discharged, which may not be the same due to losses), minimum state of charge of the battery (e.g., battery capacity floor to have a reserve amount of charge left), etc. Other parameters may be included.

[0129] The aforementioned parameters are placed into a model, where the model is run to determine, given the consumption and solar predictions for a time period of interest (e.g., tomorrow), the various energy storage device metadata described above, and any rate plans (e.g., tariffs or time of use plans), a variety of candidate battery control plans or scenarios that are each associated with a corresponding cost or value to the customer (e.g., homeowner of the site), etc. The scenarios include different ways in which the battery is controlled, where the cost of a given combination of battery control actions is determined. The risk profile of the site (e.g., amount of variability in the solar production and energy consumption forecasts upon which the battery control plans rely) is also taken into account in the modeling, as will be described in further detail below.

[0130] A control plan is then selected, and a state of charge of the battery over the time period of interest is generated.

[0131] For example, the desired or optimal state of charge over time that solves the objective function in a manner that reduces cost or maximizes value is selected. The battery is then controlled accordingly to effect the start of the charge plan or trace.

[0132] The following are further embodiments of control plan optimization, as well as determining which candidate control plan to select.

[0133] In some embodiments, machine learning predictions of production and consumption are generated, as described above, which form, for example, probability distributions. In some embodiments, N random walks are generated within that probability distribution, to create a range of future scenarios of production and consumption in kW over time. (In various embodiments, and as will be described in further detail below, N may be 3, or 11 , or any odd number). One example random walk that is taken is used to determine the energy consumption at points or steps in time over a time period, where the amount of energy consumed is based on the probability distribution determined for the energy consumption. Many random walks of the energy consumption may be taken over the same time period to generate different energy consumption traces / curves. Similar random walks are taken based on the probability distribution for solar / energy production. These production and consumption scenarios are then provided as input to an optimizer, which in one embodiment utilizes linear programming.

[0134] In some embodiments, the optimizer is provided an objective function (e.g., to minimize cost over some specified period of time, or minimize carbon, etc.), and constraints (e.g. device criteria such as max kW charge / discharge, kWh of energy capacity, device roundtrip efficiency). In some embodiments, the optimizer is configured to deliver an optimal power trace for the battery inverter in kW over time — that is, whether the battery is charging or discharging at a given point in time, in order to achieve the objective function.

[0135] At this point, an optimized battery power trace is determined for each of the production / consumption scenarios that were fed into the optimizer.

[0136] In some embodiments, a process referred to herein as “intent generation” is performed. In some embodiments, intents are instructions that are provided to the battery over the course of the day to achieve the optimal result. In some embodiments, the generation of final intents is determined via a voting system that utilizes a range of heuristics (e.g., if / then statements).

[0137] In some embodiments, each of the aforementioned optimized power traces that are outputted from the optimizer act as candidates. For each time period (e.g., 30 minutes, one hour, or any granularity as appropriate), the candidate plans are evaluated to determine what would occur over this time period. For example, if across three candidate plans there are two charge actions and one discharge action for that period, the optimal plan will include an instruction to “charge” over that period. Here, the majority of candidate plans call to perform charging during the time period (e.g., 30 minute chunk) under evaluation. If there is a split vote (e.g., one candidate plan is to charge during that period, a second candidate plan is to idle during that period, and a third candidate plan is to discharge during that period) or no clear result, rules are applied to resolve those. For example, odd numbers of candidate plans are generated to avoid split or inconclusive decisions (e.g., to avoid a “2 votes to charge, 2 votes to discharge” situation). The selection engine then aggregates (e.g., adds up) the “winning vote” for each time period, and this constitutes the optimal plan for that day (or any other range of time under consideration, as appropriate).

[0138] FIG. 2B illustrates an embodiment of a battery storage state of charge according to a battery control plan. In this example, a plot of the state of charge of the battery over a 48-hour period is shown at 206. The control plan is determined based on the statistical forecasts of solar production and energy consumption shown in FIG. 2A (and also provided in the plot above SoC curve 206 in the example of FIG. 2B). In some embodiments, the control plan is also generated based on the cost of importing and / or exporting energy to the grid (e.g., in order to optimize for cost savings as one possible goal). In some embodiments, a graph illustrating the cost of importing and exporting energy to the grid over time may also be presented.

[0139] For example, if, for a given period, the household consumption is to be covered, the traces illustrate the expected solar generation and the planned state of charge of the battery. As the battery starts to cover the evening peaking of solar generation, the battery begins to discharge. In some regions, the batteries may be charged from the grid, in addition to solar. As one example, the optimization engine described herein makes a decision, based on a rate plan (which indicates the price of grid power at various times of day), and the predictions of solar consumption and energy consumption, whether there will be sufficient solar power to charge up the battery storage to a sufficient level to cover the expected consumption at the evening peak. For example, suppose that it is predicted that tomorrow, the solar excess is 5 kWh, but that tomorrow, during the expensive period of grid energy, there will be 7 kWh of energy consumption. That is, if left to charging only via the sun, the battery will not have enough energy stored to cover the evening peak. In this case, the battery is charged during a cheaper period (e.g., midnight, according to an example rate plan), where the battery is charged up to 78% in this example (and not fully) where the remaining 22% is to be filled up by the expected or predicted amount of solar power to be generated, so that the battery will have enough energy stored by the time the peak energy consumption occurs to match the 7 kWh expected consumption. As another example, if the energy consumption prediction indicates that for a given 30 minute block of time (e.g., between 4 PM-4:30 PM), that there will be a 1.2 kW increase (e.g., because the user typically turns on the oven at that time on weekdays), a plan is generated that takes that into account.

[0140] As shown in this example, the consumption and solar generation prediction are used to guide the optimization for generating a device control plan. For example, the predictions are generated to be used as inputs to generate an optimal control plan for the house. The optimal control plan is determined to optimize against one or more goals, such as avoiding using grid power during a period where it is expensive. In some embodiments, the genetic algorithm is configured to optimize for maximum value delivery against the metric that the algorithm is optimizing against, which may be economic (e.g., benefit of the utility against wholesale energy costs, against the tariff for the customer) and / or environmental (e.g., optimizing against grid carbon).

[0141] As described above, device metadata (e.g., minimum SoC), predictions of solar power generation, predictions of energy consumption, rate plans, etc. are provided to a solver, where the objection function of the solver is to minimize cost. The solver generates a variety of candidate scenarios on how to control the battery to minimize that cost.

[0142] The most optimal option or control plan is selected. For example, the most optimal control plan is selected based on a defined risk parameter or other metrics, which in various embodiments includes factors such as the season, market conditions, or customer preferences. In some embodiments, the plan is converted into what are referred to herein as “intents.” In some embodiments, intents are instructions that are provided to the battery to control its action (e.g., provided to a home energy management system 156 of FIG. 1 B, which is configured to control cycling of the battery according to the intents). In the example of FIG. 2B, the intents are shown as vertical lines (i.e., the lines correspond to points in time at which the battery will be instructed to charge or discharge according to the specification of the intents). The intents are issued to achieve the desired state of charge profile shown in the bottom SoC graph 206 of FIG. 2B. In various embodiments, the intents include instructions for when to charge and / or discharge the battery, when to keep it idle, the power level at which the battery is charged and / or discharged, the duration of the charge and / or discharge cycle, etc.

[0143] FIG. 2C illustrates an embodiment of an intent. The intent 208 shown corresponds to one of the vertical intent lines shown in the forecast plots of FIG. 2A. In this example, the intent, at 3 PM, is to charge at a power level of 3 kW, where the minimum state of charge is 12%, and the maximum state of charge is 88%. If the power parameter were negative in value, then this would indicate an intent to discharge. The state of charge values indicate either discharging until the minimum of the intent is reached, or charging until the maximum of the intent is reached. This is an example of an instruction that is sent to the controller of the battery. The set of intents in effect forms an itinerary for the battery for the day.

[0144] FIG. 2D illustrates an embodiment of an intent. In this example, a “balance” intent is shown at 210. In this example, “balance” is a mode in which, for this time period, if there is more solar than there is more consumption, then that excess solar is to be used to charge up the battery, where the battery is charged until it reaches the SoC max limit, which is set to 100% in this example.

[0145] The opposite of this is reflected in an intent that is sent when a consumption peak is forecasted to occur, where an intent is sent to discharge the battery to meet excess consumption until a particular state of charge (e.g., minimum state of charge). An example of such an intent is shown in the example of FIG. 2B. In this example, the intent 212 is still balanced, but with a minimum state of charge of 12%.

[0146] In this example, based on this intent, if there is more consumption than there is solar (which is predicted in the evening), the battery is discharged to cover the consumption until it reaches the minimum, at which point it will hold itself. Intents are sent to charge the battery at later times.

[0147] In some embodiments, an intent is based on the device being controlled and the local modes it has available (where, for example, “balance” is an example of an available local mode that the battery storage device can be instructed to operate in).

[0148] In some embodiments, the intent includes a minimum and / or maximum SoC band. The intent also includes an instruction to either charge or discharge. The instruction to charge or discharge may be specified according to a particular setpoint (e.g., power level). Other examples of parameters specified in intents that are sent to inverters / home sites include minimum / maximum power band, start and end times, etc. In some embodiments, such parameters may be used to smooth the execution of control plans. In some embodiments, a mode of operation is also included in the intent. The “balance” mode described above is an example of such a mode, in which the battery is controlled according to a mode in which self-consumption is desired (i.e., minimizing usage of the grid), where excess solar power is used to charge the battery, and consumption excess (not covered by solar) is covered by the battery. Other examples of modes in intents include prioritizing charging of the battery via solar, where for any solar power that is being generated, it is provided directly to the battery first (rather than other household loads) that is, charging of the battery up to 100% via solar is prioritized over providing power for other loads (which may then instead be provided power via the grid). If a mode is specified in the intent, the battery is instructed to operate within that mode. For example, if an intent indicates that the battery is to be in balance mode between 9 AM-12 PM, then the battery is controlled to take any excess solar and use it to charge the battery during that time period. In this mode, if at any point consumption is more than what is being generated by the solar panels, then the battery is discharged to meet that need.

[0149] Other examples of modes include half balance modes. For example, in one embodiment of a half balance mode, the battery is charged from excess solar, but the battery is not discharged to supply power if consumption exceeds what can be sourced from the PV panels (and instead grid power is used to supplement the solar panels, rather than the battery). On the opposite side, the battery may be placed in a mode to discharge the battery if there is any excess consumption (consumption not met by solar), but to not fill the battery with excess solar. The different modes may be used to facilitate achieving of different targets.

[0150] As described above, in some embodiments, the device control plan optimization engine 122 determines an optimal control plan for a site and transmits the device control plan to a site for execution.

[0151] In some embodiments, two days of forecasts and control plans are generated and provided to a device at a time. In the example of FIG. 2B, two days (48 hours) of predictions and control plans are shown. As shown in the examples of FIG. 2C-2E, the intents generated over the span of the 48-hour period are different, reflecting the different predictions of solar generation and energy consumption for those two days (where the predictions for the first 24 hours are not necessarily the same as the predictions for the second 24 hours). For example, one day may have a large amount of sun, while the next day may have little sun. Depending on the amount of sun (and solar production) forecasted for the next day, control plans for the night before may include completely different instructions / intents. In some embodiments, the examples of FIGS. 2A — 2E illustrate embodiments of interfaces for providing information pertaining to the state of an energy system at a site. For example, the information may be provided via interfaces presented to homeowners. As another example, the interfaces are a partner-side interface (e.g., provided to installers or operators that monitor the performance of energy system installations). For example, the interfaces may be used to facilitate customer support. For example, if a homeowner requests an explanation as to why their battery is operating in a certain way, a customer support representative can use the interfaces to determine what predictions were being made to generate the battery control plan, and provide an explanation to the homeowner accordingly.

[0152] In some embodiments, a 48-hour plan (which includes intents executed over the next 48 hours) is sent to the energy storage device. The battery storage control plan is stored locally at the device. In some embodiments, the 48-hour plans are sent periodically, such as every 24 hours.

[0153] Other durations of plans may be sent at various other frequencies. For example, rather than planning on a daily or 48-hour period, a longer term optimization may be performed (e.g., over the course of a year). In other embodiments, the intents are sent as instructions that are executed immediately upon receipt (and are not necessarily locally stored for future execution).

[0154] The overlap in plan length (48 hours) and frequency (where, for example, a new 48-hour plan is generated and delivered every 24 hours) allows for a rolling window of plans. This has various benefits. For example, with respect to redundancy and reliability, even if a site loses Internet for a day, the device will still have a day of smart planning available. In some embodiments, if after 48 hours, there is still not Internet (and a new plan cannot be sent from the cloud entity to the device), then a default control plan is utilized.

[0155] Accounting for Variability in Generating Device Control Plans.

[0156] As described above, device control plans are generated based on forecasts of solar production and energy consumption. The forecasts themselves are predictions with variability. For example, accurate predictions may be more reliably generated if a customer at a home site has a regular pattern of behaviour. However, users may vary and have random spikes in behaviour. Similarly, there may be varying amounts of variation in solar production from day to day. If the energy storage control plan is optimized based on inconsistent data, this may result in less optimized usage of resources. In some embodiments, the optimizations described herein (e.g., to generate an optimal battery control plan) are weighted based on how aberrant a household’s consumption behaviour is.

[0157] In some embodiments, to prevent overfitting of device control plans to forecasts, the process of generating device control plans takes into account the statistical variability of the solar production and energy production forecasts when determining the optimal control plan to provide as output.

[0158] For example, as described above, in some embodiments, the output of the forecasts is a probability distribution. Statistical measures such as 10th percentile, median (50th percentile), and 90th percentile values are determined.

[0159] In some embodiments, the values of the plan generated at the median are evaluated, as well as the difference in value between the 10th and 90th percentile outputs. If there is a large difference in the 10th and 90th percentile boundaries of the plan, this indicates that there is a higher likelihood of over-fitting.

[0160] For example, if there is a large difference between the 10% and 90% bounds, then this feeds back into the decision of how a prediction is used, where a more risk averse mode is entered, in which more conservative assumptions are made. For example, the control plan optimization algorithm is configured to handle the spread in predictions.

[0161] In some embodiments, a greater amount of deviation in the prediction is treated as a higher risk when selecting a control plan, where the higher risk is translated as a higher cost. In some embodiments, a wide probability distribution reflects a likelihood of a particular future occurring, and optimization is performed against these scenarios, which is one example location of where the risk is captured. A more aggressive or pessimistic approach may also be taken by adjusting how tightly optimization plans are fitted to a particular exact eventuality occurring.

[0162] In some embodiments, the optimization algorithm used to generate the device control plan performs a random walk to create a range of candidate future scenarios within the probability distributions of the solar production and energy consumption forecasts, evaluates the max and min predicted values (10th and 90th percentile values), and, given the various forecasts and optimization goals described above, generates a plan for the day that is within the minimum and maximum. Where there is a larger deviation between the min and max (where the risk is reflected in the tightness or closeness of the 10th, median, and 90th percentile traces, and a larger gap or area between the min and max is indicative of larger risk of inaccuracy), the optimization is performed according to a tighter region around the median.

[0163] If there is a smaller gap, then more variability between the 10th and 90th percentile predictions is allowed when performing the optimization.

[0164] As part of the optimization process, various candidate device control plans are generated. In some embodiments, if large variation is observed for a candidate device control plan (e.g., the difference in savings there would be if the min and max forecasts were what occurred in reality), then that plan is downgraded. In this way, uncertainty is dealt with and overfitting to a predicted reality (that might not take place) is avoided. The following are further details regarding determining an optimal device control plan. For example, a random walk is taken within the prediction traces to generate a multitude of scenarios. In some embodiments, a scenario corresponds to an SoC curve for the battery (e.g., SoC level at various times during the period over which the plan is constructed). The cost of each scenario is determined. For example, if the scenario involves increasing the battery level by charging using grid power in the middle of the night (where the prediction indicates that there will be no solar), then a rate plan is used to determine the cost of that battery charging using grid energy. The various costs to the homeowner that will be incurred as a result of having the battery charged or discharged to certain levels at certain times are determined (which may in turn affect how energy is sourced to meet predicted consumption). Different scenarios will have different costs. For example, one scenario may involve charging a battery up to 100% overnight, without having solar power to top up the battery. There is a cost to fully charge the battery from the grid overnight. Another scenario may involve no charging overnight, purely relying on solar production. However, based on the predictions of energy consumption and solar generation, this may result in there being insufficient renewable sources of energy to cover the predicted energy consumption at the evening peak, in which case there will be a cost to use the grid to make up the difference in energy to support for the predicted consumption. Yet another example scenario may involve charging the battery to an intermediate amount overnight, such as 30%.

[0165] The costs of various predicted scenarios are determined. The various candidate predicted scenarios are ranked or prioritized based on cost (e.g., to the homeowner), where lower cost plans are ranked higher. In some embodiments, the candidate scenarios are also ranked according to a risk factor, where the risk factor is a measure that is indicative of the amount of variability in the solar production and / or energy consumption forecasts upon which the optimal control plan is determined.

[0166] For example, the greater the spread in the forecasts (e.g., the greater the difference between the min and max forecasts), the higher the risk factor, as the greater variability in the forecasts will translate to greater variability in the predicted costs of the scenarios. For example, consider solar production in the middle of the night. The 10th and 90th percentile predictions for solar production will be in a tight band around the median (which for example is OW for that period of time in this example). That is, the real solar production is unlikely to deviate much from the median. If the battery is to be charged at this period, then it is almost certain that any charging will be sourced from the grid (and none from solar), and the cost to charge the battery storage during the time frame can be reliably computed as the cost to draw power from the grid, without much if any expected deviation or variability. This provides a level of certainty when generating the optimal control plan for the battery storage (or EV charger, HVAC, or other energy resource). On the other hand, consider the prediction of daytime solar production. Because of the variability of various factors such as cloud cover, shading, etc., the actual amount of solar production could be in a wide range, as reflected in the forecasted solar production plots of FIG. 2A.

[0167] The actual amount of solar power generation may deviate substantially from the median, and thus, if a control plan were generated based solely on the median for this period (and relying on the median amount of solar production), then the control plan may not be optimal in reality. For example, if the 10th percentile value is significantly lower than the median production value, there is a relatively high likelihood that solar production could be far below the median value, in which case much more grid power would be needed, driving up the cost (as compared to if the median value had been solely relied upon as the basis for device control planning). If the 90th percentile value is much greater than the median, then there is a high likelihood that solar production would be far above the median value, in which case much less grid power would actually be needed, and thus there would be a lower actual cost for charging the battery storage (as compared to if the median value were relied upon). For example, suppose a candidate control plan is optimal if the actual solar production and energy consumption track closely with the median forecasted solar production and energy consumption, and is sub-optimal if the reality deviates away from the median values. If the 10th and 90th percentile bounds of the forecasts are close around the median, then this actual tracking of reality to the median is likely to occur. However, if there is a large gap between the 10th and 90th percentile values for the forecasts, then this is an indication that it is unlikely that reality will track the median value, and the control plan is unlikely to be optimal in reality / practice.

[0168] Based on the ranking based on cost and / or risk profiling, a candidate scenario is selected as an optimal energy storage device control plan for dictating the behaviour of a battery storage device at a site. In some embodiments, that plan is converted into a plan executor that takes the desired energy storage SoC curve (e.g., state of charge of the battery storage over time) and generates instructions (e.g., intents described above) to achieve those planned battery SoCs at the planned times of the selected plan. For example, if the SoC is to be increased from X amount to Y amount Within a certain period of time (e.g., half an hour), an intent is generated to charge the battery at a certain rate (e.g., power level) for a duration of time, until the maximum specified SoC for that intent (where the power level of the charging is determined based on the amount of time that charging is allowed to occur, and the amount to be charged by, which is determined based on the upper SoC limit).

[0169] Thus, as described above, in some embodiments, for a given plan, a value (e.g., saving or lowest cost) for the household is generated. The genetic algorithm generates numerous (e.g., thousands of) plans. Based on the savings and the variability of the candidate plans, an optimal plan for the household or site is selected.

[0170] In some embodiments, a list is maintained of sites with more than a threshold amount of variability. For such sites, a structurally more risk-averse approach is taken. In other embodiments, the variability of a site is determined from the probability distributions generated for the site, and a more risk-averse approach is taken when the variability of the site, based on the probability distributions, is greater than a threshold. In some embodiments, the variability of a site is also determined based on a comparison of actual consumption and / or solar generation readings against predictions to determine how closely (or not) reality tracks with the predictions being made. Such information is then used as training to improve the model going forward.

[0171] Re-Triggering Control Plan Generation / Optimization

[0172] As described above, in some embodiments, on a periodic basis (e.g., every 24 hours) a new control plan is generated (e.g., for the next 48-hour period). In some embodiments, re-planning (e.g., re — generation of a control plan) is triggered in reaction to what is observed in the real world. As described above, based on the predictions of solar generation and consumption, an expected state of charge trace over the course of a day is generated. In some embodiments, the expected state of charge trace over the course of the day is associated with a threshold or performance envelope. In some embodiments, if the threshold around the expected state of charge trace is exceeded (where the actual or real SoC deviates from the expected or planned SoC by more than a threshold amount), then in some embodiments, site-level optimization engine (124) is configured to trigger a re-plan for the site, in which new predictions and a new control plan are generated for that day.

[0173] For example, retriggering is performed based on SoC deviation (e.g., a threshold number of percentage points above or below an expected SoC percentage). In some embodiments, planning is also retriggered when an EV is plugged in, or when a flex event is triggered. Planning may also be retriggered to perform rebalancing, which in some embodiments occurs when an expected overall power delivery for a grid service is not met, and additional energy devices need to be dispatched in order to meet that overall power import or export requirement.

[0174] For example, with new measured data (e.g., additional collected metering data of PV production and site energy consumption), new machine learning predictions of solar generation and energy consumption are made. A new control plan is then made based on the new predictions.

[0175] In some embodiments, by default, plans are sent each night at a predetermined time (e.g., 10 PM, or any other time as appropriate). If a deviation (that exceeds a threshold boundary condition) with respect to the planned SoC occurs in the middle of the day, suppose at 2 PM, an ad-hoc re-plan or refresh is triggered, and a new plan is sent. In some embodiments, the ad-hoc plan that is generated is another 48-hour plan, just as with a regularly scheduled plan (e.g., has a same planned period). In some embodiments, the regular default planning is still also performed, where another plan will be generated and sent at the predetermined time (e.g., 10 PM in this example). In some embodiments, the local battery controller (e. g., energy management system 156) is configured to execute, at a given time, the most recently received intent that is available for that moment.

[0176] FIG. 3 illustrates an embodiment of a process for generating an energy storage device control plan. In some embodiments, process 300 is executed by device control plan optimization engine 122. In some embodiments, the energy storage device control plan is used to selectively store and discharge energy at a plurality of sites. The process begins at 302, when a plurality of energy (e.g., solar) generation scenarios is generated for each site in a plurality of sites, the plurality of energy generation scenarios having different energy generation scenario likelihoods. At 304, a plurality of usage scenarios is generated for each site having different usage scenario likelihoods. In some embodiments, the energy (e.g., solar) generation and usage scenarios are determined by performing random walks according to probability distributions generated for the solar generation and consumption, further details of which are described above.

[0177] At 306, a plan to selectively store energy in an energy storage device (e.g., battery or fuel cell) or discharge energy from the energy storage device is generated based at least in part on a plurality of optimization criteria and on the plurality of energy generation scenarios and usage scenarios.

[0178] In some embodiments, the optimization criteria include residential-side criteria (e.g., behind-the-meter criteria) and grid-side criteria (e.g., front-of-meter criteria, such as those relating to virtual power plants (VPP), further details of which are described below). In other embodiments, the plurality of energy generation scenarios and usage scenarios are used to determine how to control consumption of energy by loads (e.g., EVs, EV chargers, HVAC units, etc.). This includes, for example, temporally shifting usage of such controllable loads, such as controlling when EVs are charged (which may also be based on the predicted time at which they will be connected / disconnected), controlling when HVAC units such as air conditioning units or heaters are turned on (e.g., when there is predicted to be excess solar), etc. In this way, energy devices in a home are optimally controlled in order to achieve various functions or goals, such as cost saving, grid service delivery, etc.

[0179] FIG. 4 illustrates an embodiment of a process for executing an energy storage device control plan. In some embodiments, process 400 is executed by an energy management system such as energy management system 156.

[0180] The process begins at 402 when a plan to selectively store energy in an energy storage device or discharge energy from the energy storage device is received. In some embodiments, the plan is generated using process 300 of FIG. 3. In some embodiments, the plan is received from a remote entity (e.g., planning system 106) configured to generate the plan. In some embodiments, the plan is received at a controller configured to control behaviour of the battery. At 404, energy is selectively stored in the energy storage device or discharged from the energy storage device according to the received plan. For example, the energy storage device is cycled according to the intents in the received control plan.

[0181] EV Optimization

[0182] The number of electric vehicles (EVs) is increasing. EVs typically have large batteries, and significant amounts of power may be needed to charge up the energy storage of an EV, as compared to other loads in a household.

[0183] In some embodiments, as part of predicting consumption by evaluating historical customer behaviour over time, the energy provisioning planning system performs a machine learning prediction to predict when an EV at a site will be charged (e.g., plugged or unplugged). This EV related information is also included in training data used to train the machine learning model to determine the energy consumption forecast that is in turn used to determine an optimal control plan for controlling the home’s energy storage device. For example, optimization between the energy storage system (ESS) and the EV and the household loads is performed to prevent competing for solar power generated by PV panels.

[0184] In some embodiments, EV connection and disconnection events are modelled independently and are then used to sample different scenarios for the planner to optimize within a min-max fashion. For example, optimization is performed in such a way that the worst case scenario has the minimal cost possible.

[0185] For example, goals may be set to reduce grid import during expensive periods, but also keep the battery full, all while needing to fulfil charging of the EV (which may be a significant amount of energy consumption to be covered). As one example, a control plan is selected that is optimal in terms of minimizing cost to the homeowner While achieving the desired goals. As described above, the selected control plan will be a function of the predicted amount of solar generation over time, a rate plan, etc., where an optimal control plan that most efficiently utilizes the “free” solar energy and minimizes any grid charging while fulfilling consumption needs is met. In this way, management of charging of the battery and of the EV is optimized in a manner that minimizes cost.

[0186] As one example, consider a determination of an optimal time / manner in which to charge an electric vehicle (EV). In this example, using embodiments of the techniques described herein, the planning system is configured to determine what the sun will be doing throughout the day. The system also determines a predicted non-EV consumption. A user may indicate (e.g., via a user interface such as a mobile app, further embodiments of which are described below) that they need a certain amount of range (e.g., 80 miles) by a certain time (e.g., 2 PM, because they are picking up their children from school). In some embodiments, the EV usage is predicted based on analysis of historical consumption. The various goals and predictions are taken into account in order to generate an optimal plan for utilizing the battery storage device, EV, or HVAC (where there may also be a higher level optimization goal of making the most amount of money given a tariff or metering plan, or to minimize the amount of grid usage for environmental reasons).

[0187] While in this example, the site-level optimization (to reduce the overall cost to the customer or owner of the house) is configured to take into account a load such as an EV, other types of load devices (which may be controllable) may also be taken into account when performing site-level optimization, such as heat pumps, and their effects (e.g., on heating and cooling, which may affect the consumption and usage of heaters and air conditioning units).

[0188] Front / Behind the Meter Optimization

[0189] In the above examples, the energy provisioning planning system makes predictions and control plans on a per-site basis. In some embodiments, the individual site level predictions and energy storage device control plans are aggregated to determine aggregate predictions and control plans for optimization against goals at a multi-site level. In some embodiments, the multi-site level optimization is run along with the individual-site level optimization to determine device control plans that are optimal at both an individual site level, as well as a multi-site level (where such characterization is also referred to herein as “co-optimization”). For example, the impact of the multi-site level optimization on the individual-site level optimization is considered, and vice versa. For example, consider virtual power plants, which are a collection of home energy storage devices (e.g., batteries and EVs) that are controlled in aggregate to perform one or more actions (e.g., to supply power, effectively becoming a power plant). As described above, using the techniques described herein, behind-the-meter optimizations are performed, where the states of individual devices (energy storage) at various points in time are accurately predicted on a per-site basis. This provides a bottom-up sum of all of the various devices that may be used as a fleet in a virtual power plant context. Using these individual device level predictions, the energy provisioning planning system is able to accurately determine, for example, at 2:30 PM tomorrow, the predicted amount of energy available across the fleet of devices. Having such a granular level of prediction (on a site-level basis) provides more accurate dispatch of such energy. In this way, the larger scale (multi-site level) prediction is based on the aggregation of many smaller predictions (individual site level). This also allows various errors to cancel out, providing a high-quality result.

[0190] As described above, based on tariff or rate plans, using the site-level energy optimization techniques described above, the device control plan optimization engine determines a behind-the-meter optimization for a specific site, where based on a variety of factors and goals — such as known expensive periods of using grid energy, which are to be avoided, as well as predictions of how much solar will be generated and the amount of consumption for a site, and a prediction of the state of charge of a battery at that time — an optimal control plan for deploying the charge in the battery at a site is determined. For example, the battery control plan is generated so that if it were battery power being consumed by the household, then importing energy from the grid is avoided, yielding a cost savings.

[0191] However, suppose there is an issue with a substation. In this case, there may be a benefit to discharging the battery (as well as the batteries of other residential sites) back to the grid during such an event. In some embodiments, the energy provisioning planning system performs an optimization of whether the device should be controlled in a manner to facilitate discharging to the grid (an example of a front-of-meter optimization). In some embodiments, with respect to the control of the battery, the energy provisioning planning system compares the benefit of discharging the battery to the home loads (for self-consumption) against the benefit of discharging the battery to the grid to determine how to optimize control of the site’s battery. In this example, behind-the-meter and front-of-meter goals are being optimized concurrently, resulting in a form of co-optimization. Using the co-optimization techniques described herein, site level optimizations may be balanced against fleet-level optimizations to determine an overall optimal plan for controlling a household’s storage device. In this way, optimization is performed based on both “behind the meter” (site level) criteria and “front of meter” (fleet / grid level) criteria. Consider the following example in the context of a virtual power plant, in which a group of individual energy storage devices at various sites (e.g., homes) are aggregated into a collective or fleet of storage devices. Suppose that it is known that a certain amount of megawatts is needed to be delivered over a time period. Each of the sites that have energy storage devices that are enrolled in such a Virtual power plant service or program are evaluated against an algorithm.

[0192] For example, each site is queried to determine an energy-cost curve that indicates the amount of energy the site’s battery storage can provide, and at what price. This results in each site having a corresponding range of options for how much energy it can provide, and the cost for delivering the various amounts of energy. The energy-cost curves across a group of sites are aggregated to determine a plan for providing an aggregate amount of power at the lowest cost to meet the power delivery requirement by utilizing the various individual sites. For example, based on the aggregate analysis for satisfying the delivery requirement, it is determined that the energy storage devices at four sites should be utilized, but not two other ones, due to their power being expensive. An aggregate delivery plan (e.g., control plan instructions for discharging the energy storage devices at the selected participant sites) for a given Virtual power plant contract is then generated.

[0193] In the above, the lowest cost is not necessarily the cost to the grid for purchasing the power from the sites. In some embodiments, the cost is the opportunity cost to the sites for deviating from the site-specific control plan determined for them. The following are examples of determining an opportunity cost for each of the participants in the delivery plan. For example, as described above, for each site, a site-level “behind the meter” control plan is determined that optimizes for goals such as saving costs.

[0194] When a site is involved in a VPP delivery control plan, their energy storage device is being requested to operate in a manner that deviates from the optimal site-level “behind the meter” control plan. Because the homeowner is deviating from their site level control plan (which has been calculated to be the optimal plan from, for example, a cost saving perspective), the homeowner’s cost will increase, because, for example, they will not have as much energy stored in their battery in the evening to help them save on their bill. In some embodiments, the opportunity cost for a site is determined as the difference between the amount of money that would be saved if the site-level optimal control plan had been utilized, versus the amount that would be saved if the energy storage device were discharged as part of the VPP service. For example, suppose that the behind-the-meter control plan saves the customer $3 on that particular day. Because the customer was dispatched for the VPP service, they now have one kWh less in energy in the storage device. In this case, suppose that the savings is now $2.50.

[0195] In this case, the cost of them providing that kWh is 50 cents.

[0196] In this example, the opportunity cost is the estimated lost savings from performing the VPP service, where the cost is a decrease in the amount saved.

[0197] How much a user is paid is dependent on a business model. In some embodiments, the opportunity cost is not directly used to determine a control strategy, but it may be generated as a potential input to an optimization algorithm.

[0198] The opportunity cost can facilitate up front “cost to deliver” estimates of a long term contract, or for short term trading. The opportunity cost may also be ignored or unused altogether, for example because the customer does not own the asset (e.g., battery storage), or has an arrangement with their energy company / provider, etc.

[0199] In some embodiments, the selected participants in the delivery control plan are determined as those that have the lowest summed opportunity costs. Here, in embodiments of the co-optimization techniques described herein, the site-level opportunity cost to participating in the front-of meter control plan is determined (where the opportunity cost is determined as the deviation relative to the savings that would be realized if the site utilized the behind-the-meter control plan). Using the techniques described herein, the energy provisioning planning system determines an optimal plan for providing, in aggregate, a requested amount of front-of-meter energy, at the lowest opportunity cost to those customers. An assessment is made in terms of how to combine slivers of provisioning of value and opportunity cost to the customer behind-the-meter to determine the total cost for delivery of the VPP service.

[0200] In this example, co-optimization is performed, as there are two optimizations occurring, where both goals (VPP goal and behind-the-meter site specific savings goal) are treated as objective functions to meet the power requirement at the lowest cost. Using the cooptimization techniques described herein, the cost of delivery of the VPP request is a function of the opportunity cost to the site customer. The opportunity cost is based on the site-level behind-the-meter control plan that is generated based on the solar generation and consumption predictions / forecasts described above. The opportunity cost is then weighed against when performing the optimization to satisfy VPP power delivery with the goal of lowest opportunity cost for site customers. That is, the optimization of VPP power delivery is based on the site level impacts, which are in turn based on site-level predictions of solar power and consumption, as well as site-level control plans that maximize benefits such as savings to homeowners.

[0201] Using the site-level control plans determined based on the site-level forecasts of solar power generation and energy consumption, the planning system 106 is able to determine the state of the fleet of energy storage devices in an accurate manner. Having sitespecific state of charge information for individual devices allows for granular generation of control plans for satisfying VPP power delivery requests. For example, without visibility into the SoC of sites, guesses as to the state of charge of devices are made (e.g., an average SoC for each device is made). In order to account for buffer, more devices than may actually be needed are dispatched to participate in a VPP power delivery event. However, with the more accurate View of each site, as described herein, fewer devices can be dispatched, which brings down the cost of the service to both the purchaser of the power, as well as to site customers.

[0202] That is, in some embodiments, using the techniques described herein, the energy provisioning planning system determines, for each storage device in a fleet of storage devices, the SoC at any given point in time (using the behind-the-meter predictions described above). To satisfy a VPP power delivery request, an amount of aggregate power to be delivered is specified. Each energy storage device is queried to determine how much energy it can provide and at what cost. A range of options is provided for each site. The various options for the site are used to determine what the SoC of the device will be at the time of the VPP power delivery event. The opportunity cost for each site is determined. An optimization is performed to determine which subset of devices should be chosen and dispatched together to provide the requested amount of VPP power. When the individual devices are dispatched, the decision to dispatch the devices is based on the cost for a site’s customer (e.g., savings lost) to deviate from their behind-the-meter control plan, as compared to what the customer gains from participating in the VPP dispatch (e.g., what the customer will be compensated for participating in the VPP power delivery).

[0203] Using the techniques described herein, a comparison is made between the behind-the-meter cost and the front-of-meter benefit.

[0204] The co-optimization may be performed dynamically. For example, the co-optimization may be performed every half hour, or any other time-driven basis as desired. The co-optimization may also be triggered whenever there is demand for VPP power delivery.

[0205] In some embodiments, front of meter optimization also considers device efficiency in allocation control plans to devices within a fleet, for example so that a device discharges at a higher kW for a shorter period of time and is then seamlessly followed by other devices in the fleet in a “sprint relay” fashion. This is more efficient than all the devices discharging simultaneously at a lower kW. For example, suppose that there is a demand for 10 kilowatt-hours of energy. Suppose that 10 devices are dispatched to provide the requested 10 kilowatt-hours. One way to do so would be to have each device contribute equally, where the devices operate in parallel, providing the same contribution of energy. For example, each device discharges at the same rate of 1 kW for 1 hour. However, discharging at 1 kW may not be the most efficient discharge rate. For example, it may be more efficient to discharge a battery at a higher power. In some embodiments, rather than having each device be controlled in the same way, the dispatched devices are controlled to discharge in a relay or rolling manner, where a first subset or group of devices (one or more devices) is discharged at a higher power for a shorter period of time. After this, a next group of devices is controlled to perform discharging for a next period of time. The net effect of this is that power is extracted more efficiently from the dispatched devices, while also meeting the aggregate demand for power. In this way, the virtual power plant optimization considers not only the opportunity cost for owners of the devices, but also optimizes for the operational efficiency of the devices, will continuously assess performance during an event, and if delivery drops beneath a given threshold against an expected delivery, reinforcement devices are provisioned in order to return to the expected delivery curve. For example, suppose that a certain amount of power is to be delivered, however a group of devices disappear and are no longer providing power (e. g., they have gone offline (e.g., lost communications) for some reason, reducing the number of devices providing power). In some embodiments, the amount of power being delivered by devices is continuously monitored.

[0206] If there is a deviation from an expected power delivery due to a device going offline or otherwise becoming unavailable, then the system performs rebalancing of the overall delivery, for example, by bringing online reinforcement devices to fill in the gap in power delivery.

[0207] FIG. 5 is a flow diagram illustrating an embodiment of a process for co-optimization. In some embodiments, process 500 is executed by device control plan optimization engine 122. The process begins at 502 when a request to provide an aggregate amount of power at a requested time Via a plurality of energy storage devices at a plurality of sites is received.

[0208] At 504, for each site in the plurality of sites, an opportunity cost for delivering a portion of the aggregate amount of power at the requested time, using an energy storage device at the site, is determined. In some embodiments, the opportunity cost is determined based at least in part on a deviation from an energy device control plan generated based at least in part on at least one of a forecasted solar production or a forecasted energy consumption associated with the site.

[0209] At 506, opportunity costs for the plurality of sites are used (e.g., aggregated) to determine an optimal power dispatch plan including at least a subset of the plurality of sites. Different sites in the subset of the plurality of sites are instructed to provide varying amounts of power to satisfy the request to provide the aggregate amount of power at the requested time. In some embodiments, determining the dispatch plan includes providing each site in the subset of sites a plan based on the opportunity cost determined for the site. Such opportunity costs and the request for aggregate power are examples of gridside (front-of-meter) and residential-side (behind-the-meter) criteria that are evaluated to determine a device control plan. In some embodiments, a device control plan provided to a site is executed by an energy management system (such as energy management system 156) using a process such as process 400 of FIG. 4.

[0210] In some embodiments, process 500 is an embodiment of process 300 in which both front-of-meter (grid-side) and behind-the-meter (residential-side) criteria are evaluated to determine how to control charging or discharging of energy storage devices at individual sites.

[0211] Hybrid Tariff Co-optimisation

[0212] As discussed above, the system optimises some cost function based on predictions of consumption and generation at the end site and other constraints, such as battery parameters, user preferences, etc. A possible cost function may seek to minimise the price the user pays for the power, i.e. their tariff applied to their net use of power, i.e. behind the meter benefits. Another cost function may depend on a network cost, such as the wholesale price paid by the utility company and seek to minimise this cost to minimise the price paid by the utility to supply the end user. Co-optimisation techniques described above generally discuss generating an optimum control plan to optimise behind the meter benefits, i.e. to minimise the cost to the consumer based on their tariff, and then evaluating the opportunity cost for deviating from that plan to deliver benefits in front of the meter, e.g. for the battery to participate in a virtual power plant scenario. Individual batteries may be selected to participate in the virtual power plant based on the ranking of opportunity cost and / or risk to minimise the cost to the homeowner.

[0213] However, it is still simplest to optimise based on a single cost function and modifying the system to optimise against multiple separate cost functions may be difficult and require extensive modifications of existing systems. In other words, there are technical challenges in generate a control plan for co-optimisation that optimises for an individual battery between tariff, i.e. the price at which the consumer pays for power, and wholesale prices, i.e. the price the utility pays to supply the power using conventional techniques. It might for instance be contemplated to minimise the primary cost function (i.e. the customer cost / tariff) first and then minimise the secondary cost function (e.g. the wholesale cost) applying the constraint that the primary cost did not increase. However, this may not lead to the best results. Furthermore, many existing systems are set up to have a simple Behind the Meter operation, i.e. optimising on tariff, and including another factor would mean overhauling the BTM algorithms and infrastructure.

[0214] Figures 6 to 11 illustrate a co-optimisation technique based on generating a hybrid tariff is an optimum price to be used by the optimiser in generating a charging plan based on a weighted sum of the wholesale price and customer tariff. In an embodiment, the hybrid tariff is defined by:

[0215]

[0001] optimum price = (1 -x) * customer tariff + x * wholesale price

[0216] Figures 6 to 10 shows various plots taken over a 24 hour period for various parameters involved in an optimisation process for respective different values of x, i.e. weightings of the customer tariff and wholesale price.

[0217] Figure 6 illustrates the case where the value of x =0, i.e. the hybrid tariff is based only on the customer tariff. Line 602 shows predictions of energy used at a site (in kW), i.e. by the power consuming devices at the site, and line 604 shows predictions of energy generated at a site, i.e. by PV panels or other renewable sources. As expected, the usage varies over the course of the day according to the activity of the occupants and generation peaks during the day based on the movement of the sun with some intermittency due to cloud cover, shadowing etc.

[0218] Plots 606-620 show various tariffs. Plots 606 and 608 show the prices the end user pays for consuming (i.e. using power from the grid) and generating (i.e. feeding power back to the grid) according to the tariff agreed with the utility company. In this example, the tariff is in accordance with the "Denka Jozu" tariff offered by the Tokyo Electric Power Company (TEPCO) to their customers. As can be seen, the tariff for consumption 606 has various tiers, being lowest during the night (midnight to 7am), at an intermediate "off-peak" level in the morning (7am to 10 am) and evening (5pm to midnight), and at its highest "peak" level during the day (10am to 5pm). The tariff 608 for generating power sold back to the grid is constant throughout the day. It will be appreciated that many other tariffs could be used and are applicable to the present invention.

[0219] For comparison, plot 620 shows the wholesale prices paid by the utility company during the 24 hour period. In this example, the JEPX (Japan Electric Power Exchange) wholesale day-ahead price is shown. This is published at 10am a day ahead of time. As can be seen, there is more variation in the price during the day, with the price varying in 30 minute intervals. Periods where the price is most expensive do not necessarily correlate with when the tariff price is most expensive, i.e. wholesale prices may be more expensive at night.

[0220] In the example shown, the optimisation, e.g. using a linear optimiser to minimise the cost function as described above, is based on the tariff and not on the wholesale price, i.e. x = 0. Hence, the costs 616,618 used by the optimisation are the same as the tariff plots 606,608. The optimiser generates a battery plan as shown by plot 622 in which the battery is charged during the least expensive tariff tier, i.e. during the night from midnight to 7am. In this case, charging beings as soon as the night tier tariff begin, i.e. at midnight in this example when the lowest rate begins, and continues charging until the State of Charge reaches 100% or some other predetermined maximum level selected as appropriate to cover peak-time consumption later in the day, as shown by plot 624. NB, charging could have been completed at any time during the lowest cost period with no difference to the cost, but charging as soon as possible is prioritised in this scheme. During the day, when the tariff is at its intermediate off-peak and peak level, the battery is caused to discharge during periods where the solar power generated does not cover the power consumed, i.e. to avoid buying power from the grid during the more expensive periods.

[0221] Plots 626 and 628 respectively show the aggregate costs of the wholesale cost and the tariff costs as energy is used during the day. It can be seen that the plots are closely aligned, diverging where the wholesale cost is more than the tariff cost (i.e. the morning period) and converging again where the tariff cost is greater than the wholesale cost (i.e. morning and day time). Thus, the scheme minimises the cost to the consumer, but does not minimise the cost to the utility.

[0222] Figure 7 shows the same plots as Figure 6 (with like reference numerals 7XX representing the same plots and features 6XX in Figure 6, and similarly for Figures 8 to 10). In this example, x = 100, i.e. the hybrid price 716 used in the optimisation is based only on the wholesale price 720. (The tariff for consumption and generation 706,708 being shown for comparison). As can be seen, the lowest points of the wholesale price 730 occur at different times during the day which causes the battery to charge at these times 732. Thus, at these times energy must be obtained from the grid to charge the battery and cover the household consumption in excess of the solar production. These times do not all coincide with the lowest cost, night tariff periods, meaning that the cost to the utility company is minimised, but the cost to the customer is higher than the cost shown in Figure 6 due to a proportion of the battery charging being at off peak and peak times, i.e. the increase in cost at point 780 coinciding with the charging during the most expensive tariff time.

[0223] Whilst Figures 6 and 7 show the extreme cases which equate to optimising for a single variable only, Figure 8 shows an example where x=0.1 , i.e. the hybrid tariff used in the optimisation is a weighted sum of the customer tariff and wholesale price.

[0224] In this example, the hybrid tariffs 816,818 are still heavily weighted towards the consumer tariff, but now have some variation in view of the wholesale price. NB this variation is not sufficiently heavily weighted that the hybrid price changes out of band, e.g. hybrid price during any particular tariff tier is never more expensive than in a lower tariff tier.

[0225] Thus, charging still occurs during the lowest cost night period. However rather than charging immediately, the charging takes place at the times where the wholesale price is lowest, thus leading to a reduction in the cost to the utility without increasing the cost to the consumer. Figure 9 shows an example where x = 0.4, i.e. more evenly weighted between the tariff and wholesale price. In this instance, the plots are shown over a 48 hour period. During an initial period 945, the battery is discharged to supply the household consumption (with no solar generation during this period). In the next period 950, energy is obtained from the grid to cover consumption. Next, as solar generation ramps up 960, solar is used to supply the household consumption and the battery is charged fully from the excess solar generation. Once the battery is fully charged, any excess solar production is sold back to the grid 970. As solar production tails off, the battery is used to supply the household consumption during the high tariff period.

[0226] It can be seen that there are now points, e.g. 940, where the hybrid tariff is lower during a peak period than at points during off peak periods due to the increased influence of the wholesale price in the hybrid tariff.

[0227] At yet higher values of x, e.g. x=0.8, the tariff tiers are almost completely lost so the high cost period for optimisation does not reflect the high cost period for the customer. The battery is generally encouraged to charge so that it can discharge during the higher wholesale periods. However, since export is a flat rate to the customer, and there is sufficient solar to cover consumption, this creates unnecessary cost to the customer. The result is that the battery charges during the morning period spanning off-peak and peak periods in order to discharge later in the day. It can be seen that much of the charging can be achieved via solar rather than grid with any excess solar exported back to the grid once the battery is fully charged. However, for houses with poorly provisioned or no solar, this might lead to undesirable charging in localised periods. As can be seen, the cost to the customer 926 may still be reduced overall but a considerable cost saving is made for the utility company.

[0228] Figure 10 shows a plot illustrating the cost to the customer 1010 compared with the cost to the utility company 1020 for different values of x. It can be seen that there is a cost saving to the utility in moving from x=0 to x=0.1 due to the off-peak charging now occurring during periods where wholesale price is lower. There is no cost to the customer in this. There are further small incremental benefits to utility cost in moving to x=0.2. At this point, there are further benefits to moving to x=0.4 but at increased cost to the consumer. It has been found that it is possible to provide a 4% wholesale saving without impacting the customer using x=0.2. It may be possible to increase the wholesale saving by increasing x to between 0.2 and 0.5 if appropriate incentives are given to the customer. By simulating across multiple sites and historical generation and usage data, x=0.2 (approx.) has been determined to be an optimum value. Adding the dynamic wholesale component makes the battery response more proactive, i.e. not waiting to saving SoC until the end of the tariff tier). This is beneficial when solar is greater than predicted but has minimal negative effect when solar is less.

[0229] Figure 11 shows a system 1100 for implementing the hybrid tariff charging plan. The wholesale prices 1102 and consumer tariffs 1104 are fed to the optimisation engine in a cloud server 1110 and combined with an appropriate weighting to give a hybrid tariff 1106 for the upcoming period of optimisation, e.g. 24 or 48 hours. Predictions of PV generation and consumption 1108 from the energy consumption prediction engine 118 and energy generation prediction engine (described in detail elsewhere) are also fed into the system for the upcoming period. The optimisation engine 1110 then optimises the charging plan according to the inputs and constraints of the system, as described above. Where plural candidate plans are generated for different generation and usage scenarios, a final control plan is determined, as discussed above. The plans are then dispatched to the batteries 1112 and / or loads over a network 1111 via a series of intents as discussed above.

[0230] In the example described above, the resource being controlled is a storage device, such as a domestic storage battery or an Electric Vehicle when connected to a charging station. Various constraints can be included in the optimisations, e.g. for an EV battery, the battery should be charged to a particular level at a particular time in anticipation of a journey.

[0231] The hybrid tariff optimisation approach may also be used for controlling a load, e.g. an air conditioning unit, by time shifting its energy consumption. As discussed above, in the scenario where an energy storage device is being controlled, the plan determines when the device "consumes" energy (i.e. stores energy) from the grid and / or renewable sources at the site and when it discharges. In contrast, in the scenario where consumption of a load is controlled, e.g. by time shifting, the plan more simply determines when the load is made to consume energy at the site from the grid and / or the local renewable sources based on optimising the hybrid cost function and the predictions of energy usage / generation.

[0232] In the example above, the network cost profile is a wholesale price, e.g. indicating the price the utility company must pay for an upcoming time period. Alternatively or additionally other network cost profiles could be used in forming the hybrid tariff. A possible alternative cost could be predictions of carbon intensity on the grid supply for the next day. Carbon intensity is a measure of how much CO2 emissions are produced per kilowatt hour of electricity consumed and varies by hour, day and season due to changes in demand, low carbon generation (renewables, nuclear, etc) and conventional generation. This can be forecast in advance, e.g. 24hours or 48 hours and be used instead of the wholesale price in generating the hybrid tariff based on a weighted sum of the carbon intensity and the tariff.

[0233] Similarly another possible alternative network cost could be the variable cost of settling grid imbalance. The Imbalance Price is used to settle energy imbalance volumes. At the end of a Settlement Period, a Party’s contracted (traded) volume is compared with the metered volume of energy used in the Settlement Period. If a Party is in imbalance of its contracted volume, then it will be subject to imbalance charges, the cost of which can vary per hour, day, season, etc. and can be forecast in advance, e.g. 24 hrs or 48 hrs Thus, predicted imbalance price can be used instead of wholesale price in generating the hybrid tariff based on a weighted sum of the imbalance price and the tariff.

[0234] The network cost might also be localised rather than at the level of the entire grid. For instance, in markets with locational marginal pricing, the hybrid tariff can be based on the pricing node in the transmission network that the customer is fed from.

[0235] It will be appreciated that other combinations of factors can be used in generating the hybrid tariff.

[0236] The hybrid tariff is then input to an optimisation engine (also referred to as a planner) implementing an optimisation algorithm which constructs the cost function to minimise the cost against the hybrid tariff for the optimisation engine's model of the site. This optimisation might for instance be based on linear programming, mixed integer linear programming, machine learning or other algorithms, etc. as desired.

[0237] The method can be combined with other techniques described herein, i.e. as described in relation to Figures 1 to 5, such as using statistical generation and usage scenarios and performing random walks to obtain plural possible scenarios to use in the optimisation. As described above, this allows different control plans to be evaluated and a majority vote taken as to what action is taken by the storage device / load during each time division of the time period.

[0238] Efficiency

[0239] When optimising for multiple devices on a single site it may be useful to also consider the efficiency with which the energy is delivered to the end use device, as in effect this can act as a multiplier on the cost of the energy finally utilised.

[0240] For instance, solar panels will only produce a limited amount of energy in any particular day, depending on the specification of the solar array and the amount of sunshine on that day (predicted in the algorithms discussed in this application). This solar energy can often be treated as being free, or having a low cost (what the energy company will pay for export), but it is a limited resource. On the other hand energy from the grid is for practical purposes unlimited, and just has a cost at a particular time of day.

[0241] So the algorithms advantageously need to effectively allocate this low cost limited resource to displace the highest cost of imported energy (or other value such as pushing power into a VPP service and being paid).

[0242] For example charging up a home battery from Solar energy, and then using this energy in the home to power an appliance such as a washing machine might have a calculation such as:

[0243] Solar (DC) to battery 98% efficient, energy into battery I energy out (DC) 92% efficient, DC to AC conversion 97% efficient --> 0.98*0.92*0.97 = 87% overall efficiency In contrast, charging an EV from Solar energy might have a calculation such as:

[0244] Solar (DC) to EV charger (DC) only one conversion 98.5%

[0245] So when considering if the limited energy available from solar should be used to charge the battery or the EV it is beneficial to also account for the efficiency of delivery to the end use case, which can be applied as a multiplier on the costs. This addresses the issue of optimization between the energy storage system (ESS) and the EV and the household loads is performed to prevent competing for solar power generated by PV panels discussed in the “EV optimisation" section above.

[0246] So, for instance, the predictions may indicate that there is 20kWhr of solar energy available in an upcoming 30min time period and the EV requires 10kWhr and the storage battery also has capacity for charging. Charging the EV is more efficient, so the EV gets charged with 10kWhr and the storage battery is allocated the rest.

[0247] This is important for limited resources such as solar, but the principle can be applied for other scenarios, i.e. efficiency in using grid, solar and / or battery to provide power to a controllable load, such as HVAC, charging an EV or battery, an appliance or selling power to the grid. Thus the cost function being optimised can have weightings applied according to the efficiency of the energy use.

[0248] These principles can be applied with the hybrid tariff principles described in the foregoing section or independently.

[0249] Embodiments of the present invention have been described with particular reference to the examples illustrated. However, it will be appreciated that variations and modifications may be made to the examples described within the scope of the present claims.

Claims

1. CLAIMS1. A method for controlling an energy storage device or load, comprising:3.receiving a tariff for a time period comprising plural tiers according to which the energy consumer at a site is charged for energy at different rates at different times; receiving a network cost profile for the time period;4.generating a hybrid tariff based on a weighted combination of the tariff and network cost;5.generating an energy usage prediction and, where renewable or other generation sources are available, an energy generation prediction for the time period for each site in a plurality of sites,6.generating, for the site, a plan to selectively store energy in an energy storage device at the site, discharge energy from the energy storage device, or control when a load at the given site consumes energy based at least in part on optimising the cost of energy consumed according to the hybrid tariff based on the energy usage and generation prediction.

2. The method of claim 1 , wherein the network cost is based on one or more of: wholesale price;8.carbon price; and9.imbalance settlement price.

3. The method of claim 1 or claim 2, wherein the weighted combination is according to the formula:11.hybrid tariff = (1-x) * consumer tariff + x * network cost, where the weighting factor x is in the range 0 < x <1 .

4. The method of claim 3, wherein x is between 0.05 and 0.

55. The method of claim 3, wherein x is between 0.1 and 0.

256. The method of any preceding claim, comprising simulating energy usage at at least one end site across multiple days of historical data and multiple different weighting values for the weighted combination obtain optimum weighting value.

7. The method of claim 6, wherein the optimum weighting values minimises cost to the utility without increasing cost to the consumer.

8. The method of claim 7, wherein optimum weighting values minimises cost to the utility whilst increasing cost to the customer by no more than 5%.

9. The method of any preceding claim, comprising calculating the cost saving to the utility company and the cost difference to the consumer.

10. The method of any preceding claim, wherein hybrid price during any particular tariff tier is never more expensive than in a lower tariff tier and vice versa.

11. The method of any preceding claim, wherein the network cost is normalised against the customer tariff before applying the weighting.

12. The method of any preceding claim, wherein the method is carried out separately for different end sites.

13. The method of claim 12, wherein the hybrid tariff is calculated separately for each energy consumer.

14. The method of claim 12 or claim 13, wherein the network cost profile is localised such that different network cost profiles are used for different consumers in different parts of the network.

15. The method of any of claims 12 to 14, wherein the predictions and optimisation is performed separately for different end sites.

16. The method of any preceding claim, wherein in optimising the cost, the cost is first weighted according to the efficiency of the electrical conversions involved.

17. A system, comprising:25.a processor configured to:26.receive a tariff for a time period comprising plural tiers according to which the energy consumer at a site is charged for energy at different rates at different times;27.receive a network cost profile for the time period;28.generate a hybrid tariff based on a weighted combination of the tariff and network cost; generate an energy usage prediction and, where renewable or other generation sources are available, an energy generation prediction for the time period for each site in a plurality of sites;29.generate, for the site, a plan to selectively store energy in an energy storage device at the site, discharge energy from the energy storage device, or control when a load at the given site consumes energy based at least in part on optimising the cost of energy consumed according to the hybrid tariff based on the energy usage and generation prediction.

18. The system of claim 17, comprising a communications interface arranged to dispatch a charging plan to a remote energy storage device or load.

19. The system of claim 18, comprising plural remote energy storage devices and / or loads at plural end sites in communication with the processor via a network.

20. A method for controlling an energy storage device or load, comprising:33.receiving a tariff for a time period representing the cost of energy at different times;34.generating an energy usage prediction and, where renewable or other generation sources are available, an energy generation prediction for the time period for each site in a plurality of sites,35.generating, for the site, a plan to selectively store energy in an energy storage device at the site, discharge energy from the energy storage device, or control when a load at the given site consumes energy based at least in part on optimising the cost of energy consumed according to the tariff based on the energy usage and generation prediction and weighted according to efficiency of consuming the energy.

21. A method according to claim 20, wherein plural energy storage devices or loads are controlled and the efficiencies include the efficiencies of electrical conversions required in providing solar energy to those devices or loads.

Images