Energy management systems and methods

Abstract

A smart water heating system is disclosed that uses advanced monitoring and control techniques to optimize energy usage and grid stability. Various features include real-time state of charge estimation for water heaters, a "digital twin" representation of each unit, predictive load modelling, and adaptive control algorithms. The system can operate individual heaters as distributed energy resources, or aggregate them into virtual power plants. It employs strategies such as load shifting to align with renewable energy availability, and cooperative multi-agent control to prevent demand spikes. The technology aims to improve energy efficiency, reduce costs, integrate renewable sources, and enhance grid stability while maintaining user comfort. Additional features include user analytics, comparative consumption reporting, and solutions for low-voltage environments using supplementary battery storage.

Description

ENERGY MANAGEMENT SYSTEMS AND METHODSSTATEMENT OF CORRESPONDING APPLICATIONS

[0001] This application is based on the provisional specification filed in relation to New Zealand Patent Application No. 803254 filed 30 August 2023, New Zealand Patent Application No. 803385 filed 4 September 2023, New Zealand Patent Application No. 805234 filed 6 November 2023, New Zealand Patent Application No. 808761 filed 4 March 2024, and New Zealand Patent Application No. 809562 filed 26 March 2024, the entire contents of which are incorporated herein by reference.TECHNICAL FIELD

[0002] The present disclosure relates to energy management systems and methods, more particularly implementing adaptive control of thermal energy storage devices such as electric hot water heating systems.BACKGROUND

[0003] Electric hot water storage heaters (referred to herein as water heaters) are essential components in many residential and commercial settings, providing a supply of hot water for various purposes. Traditional heating systems typically rely on mechanical thermostatic controls and single-point set temperatures to manage the heating process. Such conventional mechanical thermostatic control often leads to energy consumption that is strongly correlated to when hot water is drawn, irrespective of the stored energy of the water heater. This operation follows typical user consumption behaviour, burdening the power grid and resulting in increased electricity costs for consumers.

[0004] Additionally, conventional systems lack the intelligence to adapt to varying electricity prices or to utilize clean energy sources optimally. The operational deadband in traditional systems creates periods of no heating control until the temperature falls sufficiently for the thermostat to reactivate.

[0005] Alongside the global shift towards electric appliances and renewable energy, the existing power infrastructure faces challenges such as energy poverty and electrical network constraints. Traditional electrical systems are predominantly centralised, where large power plants, often running on fossil fuels, generate the bulk of electricity. These systems then transmit electrical energy over long distances to consumers, encountering significant transmission losses and inefficiencies. The centralised nature of such systems also means that any disruption can lead to widespread outages. Moreover, the environmental impact due to reliance on non-renewable sources is substantial, contributing to global climate change and local air pollution.

[0006] Recent trends are shifting the paradigm from centralised production to distributed energy systems, where electricity is generated closer to the point of use. Distributed Energy Resources (DERs) play a critical role in this transition, offering a more resilient, efficient, and cleaner energy system. Theseresources include a range of technologies such as photovoltaic cells, wind turbines, energy storage systems, and smart energy management systems.

[0007] While renewable energy sources are vital for a sustainable future, their intermittent nature poses significant challenges. The variability in power supply due to factors such as weather and time of day leads to difficulties in balancing supply and demand, with unpredictability requiring adaptation to rapid changes in energy production and consumption.

[0008] Flexibility in DERs is crucial for addressing the challenges of renewable energy integration and maintaining grid stability. This flexibility can be achieved through various strategies, including: Demand Response (i.e., adjusting consumption in response to supply conditions), Energy Storage (i.e., banking excess energy and releasing it as needed), and Real-Time Energy Management (i.e., optimising energy use based on current conditions).

[0009] Additionally, a particularly critical aspect of flexibility is the contribution to grid frequency response. Grid frequency reflects the balance between supply and demand, with a stable frequency indicating a balanced grid. When this balance is disrupted, rapid adjustments are needed to prevent damage to infrastructure and ensure continuous, reliable power. DERs can provide these rapid adjustments by quickly altering their output or consumption, contributing significantly to frequency regulation. This is especially crucial as the intermittent nature of renewable sources can lead to more frequent and unpredictable imbalances.

[0010] Recent innovations have significantly enhanced the viability and flexibility of DERs. Advances in battery storage technology allow for more efficient and longer-duration energy storage, while smart grid technologies enable more sophisticated and responsive energy distribution and management. Software for energy management can optimise energy use in real-time, adapting to changing conditions and user needs.

[0011] Moreover, advancements in grid frequency response technologies have allowed DERs to participate actively in frequency regulation. This includes faster and more accurate monitoring and control systems that can detect slight variations in frequency and adjust the operation of DERs in real-time to help correct these variations. This capability is paramount in a modern grid with high levels of renewable penetration, where traditional large-scale generators that once provided most of the frequency regulation are being phased out.

[0012] Despite advancements, gaps remain in technology and systems that prevent the full realisation of DER potential, particularly in the area of grid frequency response. Most existing systems are not equipped to dynamically manage the fast and precise adjustments needed for effective frequency regulation, especially in grids with high levels of intermittent renewable energy sources. There is a need for further innovation to enhance the real-time responsiveness and integration of DERs into frequency regulation mechanisms.

[0013] Aspects of the technology of the present disclosure are directed to overcoming one or more of the problems discussed above. It is an object of the present invention to address one or more of the foregoing problems or at least to provide the public with a useful choice.

[0014] Further aspects and advantages of the present disclosure will become apparent from the ensuing description which is given by way of example only.SUMMARY

[0015] Aspects of the present technology generally relate to the improvement of energy systems, particularly through management of thermal energy storage devices.

[0016] Aspects of the present technology relate to methods, devices, and systems for determination of a state of charge of one or more thermal energy storage devices.

[0017] Aspects of the present technology relate to methods, devices, and systems for determination of one or more characteristics of power supplied to one or more thermal energy storage devices.

[0018] Aspects of the present technology relate to methods, devices, and systems for operating one or more thermal energy storage devices based at least in part on a determined state of charge.

[0019] Aspects of the present technology relate to methods, devices, and systems for operating one or more thermal energy storage devices to compensate for passive energy losses.

[0020] Aspects of the present technology relate to heating element devices for thermal energy storage devices.

[0021] An aspect of the present technology relates to a method, including: receiving an indication of flow of liquid from a thermal energy storage device; determining a representation of currently stored energy by the thermal energy storage device, based at least in part on the received indication of flow; and controlling at least one heating source to heat liquid in the thermal energy storage device based at least in part on the determined representation of currently stored energy.

[0022] In examples the method may include controlling operation of the at least one heating source based at least in part on a demand profile for the thermal energy storage device.

[0023] In examples the method may include receiving at least one external signal indicative of at least one characteristic of an external operating environment influencing control of the at least one heating source.

[0024] In examples the at least one characteristic is one or more of: amount of curtailment, percentage of renewable energy generation, price, environmental conditions, time of use tariffs, amount of carbon emissions, and network constraints.

[0025] In examples the method comprises generating a digital twin of the thermal energy storage device based at least in part on the determined representation of currently stored energy, wherein control of the at least one heating source is based on at least one output from the digital twin.

[0026] In examples the thermal energy storage device is one of a plurality of thermal energy storagedevices, and the method further comprises: determining the representation of currently stored energy for each of the plurality of thermal energy storage devices; and controlling the respective heating sources of the plurality of thermal energy storage devices based on a cooperative multi-agent control approach.

[0027] In examples the indication of flow is a sensed pressure within a reservoir of the thermal energy storage device.

[0028] In examples the method includes continually determining a current value of the representation of currently stored energy, and adjusting control of the at least one heating source in response.

[0029] In examples the thermal energy storage device is an electric water heater and the at least one heating source is a resistive heating element.

[0030] An aspect of the present technology relates to a system, comprising: a thermal energy storage device; at least one sensing device configured to output an indication of flow of liquid from the thermal energy storage device; at least one controller configured to: determine a representation of currently stored energy by the thermal energy storage device, based at least in part on the received indication of flow; and control operation of at least one heating source to heat liquid in the thermal energy storage device based at least in part on the determined representation of currently stored energy.

[0031] An aspect of the present technology relates to a computer-readable storage medium configured with data and with instructions that upon execution by at least one processor will cause the at least one processor to perform a method, comprising: receiving an indication of flow of liquid from a thermal energy storage device; determining a representation of currently stored energy by the thermal energy storage device, based at least in part on the received indication of flow; and controlling operation of at least one heating source to heat liquid in the thermal energy storage device based at least in part on the determined representation of currently stored energy.

[0032] The above and other features will become apparent from the following description and the attached drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Further aspects of the present disclosure will become apparent from the following description which is given by way of example only and with reference to the accompanying drawings in which:

[0034] FIG. 1 is a schematic diagram of an exemplary system according to aspects of the present technology.

[0035] FIG. 2A is a schematic diagram of an exemplary local operating environment according to aspects of the present technology.

[0036] FIG. 2B and FIG. 2C are schematic diagrams of an exemplary water heater, including fluid inputs and outputs.

[0037] FIG. 2D and FIG. 2E are perspective views of an exemplary heating element and control deviceaccording to aspects of the present technology.

[0038] FIG. 2F and FIG. 2G illustrate an exemplary switching device and associated heatsink according to aspects of the present technology.

[0039] FIG. 2H illustrates another exemplary heating element and control device according to aspects of the present technology.

[0040] FIG. 21 illustrates another exemplary heating element and control device according to aspects of the present technology.

[0041] FIG. 2J illustrates another exemplary heating element and control device according to aspects of the present technology.

[0042] FIG. 2K illustrates another exemplary heating element and control device according to aspects of the present technology.

[0043] FIG. 3 is a plot demonstrating an exemplary pressure response to flow in a water heater.

[0044] FIG. 4A illustrates an exemplary method for determining state of charge of a water heater according to aspects of the present technology.

[0045] FIG. 4B is a plot of temperature against relative state of charge and confidence in order to determine a modified state of charge value according to aspects of the present technology.

[0046] FIG. 5A is a diagram illustrating exemplary ancillary energy data over time, which may be considered by control algorithms of the present technology.

[0047] FIG. 5B is a plot demonstrating exemplary power generation curves of non-dispatchable energy sources.

[0048] FIG. 4A illustrates an exemplary method for determining state of charge of a water heater according to aspects of the present technology.

[0049] FIG. 6A to 6C are plots illustrating projected state of charge of water heaters as determined in accordance with aspects of the present technology.

[0050] FIG. 7 is a schematic diagram of a system implementing an aggregated control methodology in accordance with aspects of the present technology.

[0051] FIG. 8A to 8C are exemplary load profiles of an exemplary water heater using different control methodologies.

[0052] FIG. 10A shows energy demands over time of a water heater under traditional thermostatic control.

[0053] FIG. 10B shows energy demands over time of a water heater under a control methodology according to aspects of the present technology.

[0054] FIG. 11A shows a typical residential demand profile for a group of water heaters under traditional control.

[0055] FIG. 11B shows a residential demand profile for a group of water heaters under a control methodology according to aspects of the present technology.

[0056] FIG. 11C shows a residential demand profile for a group of water heaters under another control methodology according to aspects of the present technology.

[0057] FIG. 12 is a schematic diagram of another exemplary local operating environment according to aspects of the present technology.

[0058] FIG. 13 illustrates an exemplary electric water heater according to another aspect of the present technology.DETAILED DESCRIPTION

[0059] Typical hot water heater control operation is based on one or more thermostat(s) controlling the heating element. The present technology provides an alternative approach to operation and control of hot water heaters based primarily on the available stored energy (referred to herein as the "state of charge"). The state of charge ("SOC") of a water heater is a representation of the current energy stored relative to the full capacity of the device. By accurately estimating the state of charge of a water heater, the present technology is enabled to, for example, improve energy usage, and / or enhance grid stability, and / or provide a control approach that aligns with prioritisation of renewable energy sources. It should be appreciated that in various aspects of the present technology a nominal full capacity (i.e., less than the maximum capacity of the water heater) may be utilized for control purposes.

[0060] Aspects of the present technology provide a methodology for generation of a continuous energy-based state representation - referred to here as an energy state "digital twin". This energy-based digital twin provides a granular representation of the energy state that enables more advanced control methodologies.

[0061] The present disclosure also details hardware for the collection of data and control of water heater(s). Various embodiments are contemplated, for example in which the hardware is provided as a retrofittable option for existing hot water heaters, or integrated into new systems. This hardware is utilised to monitor an associated water heater to capture data such as temperature, power and an indication of real time flow. In one example, a pressure sensor provides the indication of flow (as well as enabling other potential insights into the behaviour of the system).

[0062] This information allows for continuous updates to the estimation of SOC. Further, in examples historical data may be used to generate load profiles that reflect real fluctuations (e.g., based on the date, and / or time of day). In examples, such load profiles provide a practical operational minimum stored energy buffer to maintain operational service.

[0063] In aspects of the present technology, a relative stored energy metric can be generated from the current estimated SOC and projected load profiles. This relative stored energy metric provides an indication of a particular water heater's available flexibility (via its associated digital twin) in terms of its ability to defer heating (when relative energy storage is high), and the priority for heating (when relative energy storage is low).

[0064] The relative stored energy metric of a digital twin of a water heater allows adaptive control algorithms to interact with local environmental signals (for example, to maximise rooftop photovoltaic consumption, or schedule with other household loads such as an electric vehicle), and external environmental signals such as non-dispatchable renewable energy sources, cost of electricity, or network grid constraints.

[0065] The digital twin representation of water heaters also provides a means for aggregation of potentially geographically connected water heaters (e.g., at street, transformer, substation, or region levels). The present disclosure discusses leveraging the digital twin energy state to provide distributed operation in response to diverse local and / or network signals in a coordinated manner.1. System Overview

[0066] FIG. 1 illustrates an exemplary system 1000 in which aspects of the present disclosure may be provided. The system includes a local environment 2000, more particularly relating to monitoring and control of an electric hot water heater 2100 (see FIG. 2A and associated description below), referred to herein as a water heater 2100. The local environment 2000 includes local control data 2002, local control algorithm(s) 2004, power control 2006, and system and operating data 2008. While a single local environment 2000 is illustrated, it is envisaged that in examples the system 1000 will include a plurality of local environments 2000 (e.g., multiple households, each having an electric hot water heater 2100). In examples the local environments 2000 may be identified as belonging to subsets or clusters for control purposes, as will be described herein.

[0067] For a firmware and / or software (also known as a computer program) implementation, the techniques of the present disclosure may be implemented as instructions (for example, procedures, functions, and so on) that perform the functions described. It should be appreciated that the present disclosure is not described with reference to any particular programming languages, and that a variety of programming languages could be used to implement the present invention. The firmware and / or software codes may be stored in a memory, or embodied in any other processor readable medium, and executed by a processor or processors. The memory may be implemented within the processor or external to the processor. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, for example, a combination of a digital signal processor (DSP) and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. The processors may function in conjunction with servers, whether cloud based or dedicated, and network connections as known in the art.

[0068] In various embodiments, one or more cloud computing environments may be used to create, and / or deploy, and / or operate at least part of the software system that can be any form of cloud computing environment, for example: a public cloud, a private cloud, a virtual private network (VPN), asubnet, a Virtual Private Cloud (VPC), or any other cloud-based infrastructure known in the art. It should be appreciated that a service may utilize, and interface with, multiple cloud computing environments.

[0069] Data from the local environment 2000 is transmitted to an environment state system 3000, in examples implemented in a cloud computing environment. The environment state system 3000 includes a system database 3002, timeseries database 3004, user system state module 3006, and user behaviour prediction module 3008. Operation of the environment state system 3000 is discussed further below. Reference to an environment state in the context of reinforcement learning should be understood as a representation of the current environment that the agent interacts with, rather than the natural environment or natural world.

[0070] Outputs from the environment state system 3000 are input into adaptive control module 4000, in examples implemented in a cloud computing environment. Ancillary data 5000 is also input into adaptive control module 4000. Operation of the adaptive control module 4000 is discussed further below. Output(s) from the adaptive control module 4000 may be used in control of aspects of the local environment 2000.

[0071] It should be appreciated that while the environment state system 3000 and adaptive control module 4000 are illustrated as being hosted remotely in cloud computing environments, in alternative arrangements one or more of these functions may be hosted in the local environment 2000.2. Local Environment

[0072] Referring to FIG. 2A, the local environment 2000 includes a water heater 2100 - in this example in the form of an electric resistance storage water heater having upper and lower resistive heating elements 2102. While FIG. 2A illustrates an electric water heater 2100 having multiple heating elements 2102, it should be appreciated that this is not intended to be limiting to all examples of the present disclosure. For example, the water heater 2100 may have a single heating element 2102, or may have more than two heating elements 2102.

[0073] It should be appreciated that the principles described herein may be applied to other types of electric water heaters. While the present technology has been primarily described in the context of traditional hot water heaters, it is important to note that the principles and technologies described herein may be applied to hydronic heating systems and hybrid heating solutions. For example, heat pump water heaters commonly include resistive heating elements to provide a backup to heat pump operation to ensure demand can be met, which may be controlled according to the present disclosure.

[0074] Referring to FIG. 2B and FIG. 2C, the water heater 2100 includes a cold water inlet 2104 (typically disposed towards the bottom of the storage tank of the water heater 2100) and a hot water outlet 2105 (typically disposed towards the top of the storage tank of the water heater 2100). A mixing valve 2106 blends the cold water feed (also supplying cold water inlet 2104) with hot water fed from the hot water outlet 2105 to achieve a desired mixed (i.e., medium) temperature supply to the consumer via mixed water outlet 2107.

[0075] Returning to FIG. 2A, In the illustrated example the water heater 2100 includes thermostats 2110, configured to regulate heating elements 2102. In examples, each heating element 2102 may be controlled by their own thermostat 2110 in accordance with known thermostatic control principles to keep water within the water heater 2100 between limits, and typically contain a thermal cut out to protect from overheating. In examples, the control of the heating elements 2102 according to methodologies described herein may be subservient to the thermostat 2110 to leverage compliance with existing regulatory standards - however it should be appreciated that this functionality may be provided through alternative dedicated safety measures.

[0076] In this example the local environment 2000 includes a controllable switching device 2112 for controlling heating of the heating elements 2102. In examples, the switching device 2112 may be capable of high frequency switching -for example a triac, or a solid state relay - although it should be appreciated that this is not intended to be limiting to all examples of the present technology.

[0077] In examples, power delivered by power supply 2114 is monitored by power meter 2116 to determine characteristics such as current, voltage, and / or frequency.

[0078] In alternative examples, power consumption may be estimated based on known characteristics of an associated heating element 2102 (e.g., rated power and heat output) and operating time. For retrofit installations using existing heating elements 2102, these ratings may be input to the system by a user (e.g., via programming or setting via physical inputs such as a dip switch). By monitoring the duration for which a heating element 2102 is switched on, then the amount of power consumed can be accurately estimated. The duration may be obtained by the control signal and can be confirmed through recorded characteristics such as pressure and temperature.

[0079] It is envisaged that data relating to local voltage and / or frequency may be transmitted to the environment state system 3000 and / or adaptive control module 4000. This may enable, for example, geographical responses to real-time local changes where such data may otherwise be unavailable to affected parties.

[0080] One or more temperature sensors 2118 output an indication of water temperature in the water heater 2100. In examples the temperature sensor(s) 2118 may be provided to an exterior surface of a wall of the reservoir of the water heater 2100. In other examples the temperature sensor(s) 2118 may be provided within the reservoir. It is envisaged that in examples, at least one temperature sensor 2118 may be provided proximal the heating element 2102 (e.g., disposed towards the bottom of the storage tank of the water heater 2100). In examples, at least one temperature sensor 2118 may be provided towards the top of the storage tank (e.g., proximal the outlet 2105). In examples, a series of temperature sensors 2118 may be provided to provide a temperature profile through the height of the water heater 2100.

[0081] One or more flow sensors 2120 output an indication of flow of hot water from the water heater 2100. In examples, the one or more flow sensors 2120 may sense flow directly - e.g., a flow meter on aninlet of the water heater 2100. In examples the flow may be sensed on the outlet of the water heater 2100, however it is envisaged that sensing on the inlet may assist with capturing losses from thermal or pressure expansion that do not go through the outlet.

[0082] In examples, it is envisaged that a pressure sensor may be used to determine or infer a flow characteristic, with pressure within the reservoir being influenced by flow in a way that is distinguishable from fluctuations due to passive losses. In examples, a pressure sensor and temperature sensor may be provided in the same package. In examples, such a pressure and temperature sensor may be provided: proximal the top of the water heater 2100 (e.g., in the outlet piping or an alternative outlet port) to give both pressure and top temperature, proximal the heating element 2102 to give both the pressure and the bottom tank temperature, and / or on the inlet water piping to give the pressure of the water heater 2100 and the temperature of the incoming water.

[0083] In alternative examples, the one or more flow sensors 2120 may sense a different characteristic indicative of flow - e.g., a vibration and / or acoustic sensing device (such as a microphone and / or a piezoelectric vibration sensor), or temperature sensors 2118 from which a flow characteristic may be determined or inferred (as described further below).

[0084] Local controller 2200, for example a microprocessor, has memory 2202, and other components typically present in such computing environments. In the exemplary embodiment illustrated the memory stores information accessible by one or more processors of the microprocessor, the information including instructions that may be executed by the processors and data that may be retrieved, manipulated or stored by the processors. The memory 2202 may be of any suitable means known in the art, capable of storing information in a manner accessible by the processors, including a computer-readable medium, or other medium that stores data that may be read with the aid of an electronic device. The processors may be any suitable device known to a person skilled in the art. Although the processors and memory are illustrated as being within a single unit, it should be appreciated that this is not intended to be limiting, and that the functionality of each as herein described may be performed by multiple processors and memories, that may or may not be remote from each other. The instructions may include any set of instructions suitable for execution by the processor. For example, the instructions may be stored as computer code on the computer-readable medium. The instructions may be stored in any suitable computer language or format. Data may be retrieved, stored or modified by processors in accordance with the instructions. The data may also be formatted in any suitable computer readable format. Again, while the data is illustrated as being contained at a single location, it should be appreciated that this is not intended to be limiting - the data may be stored in multiple memories or locations.

[0085] In particular, the local controller 2200 implements the control algorithm of local control 2004 to control operation of the water heater 2100. In examples, the local controller 2200 may receive instructions from remote sources such as the adaptive control algorithm module 4000 (see FIG. 1). However, it is envisaged that the local control 2004 may override such instructions based on localpreferences - for example in response to voltage or frequency conditions as part of grid stabilization. Further, the control algorithm may revert to returning control to thermostats 2110, or a default control scheme, in the absence of a remote connection.

[0086] The local controller 2200 may also implement flow processing 2206 in order to determine or infer one or more characteristics of flow of hot water from the water heater. Determination of flow based on pressure sensing is discussed further below. Again, it should be appreciated that alternative means for determining flow are contemplated. For example, artificial intelligence may be applied to data from at least the vibration and / or acoustic sensing device 2120 - such as machine learning pattern recognition - to determine one or more of: a flow state (e.g., flow or no-flow), and a qualitative indication of flow intensity (e.g., low, medium, or high). While this determination may be performed remotely, it is envisaged that determining flow characteristics locally may assist with timely responses to changing conditions locally. It is envisaged that the artificial intelligence may be capable of distinguishing flow due user demand from anomalies such as the relief valve passing.

[0087] The local controller 2200 may include a communications module for communication with the environment state system 3000 and adaptive control module 4000. In the illustrated example a wireless modem 2208 and antenna 2210 allow for wireless communication over one or more networks (e.g. a WiFi connection or a cellular network connection). In alternative examples the local controller may be provided with a wired communication connection.

[0088] In examples, the local environment 2000 may include a GPS device to enable determination of the geographical location of the local environment 2000. It is also envisaged that the GPS device may be used to provide a timing reference in determination of frequency of the electrical supply by the local controller 2200.2.1 Hardware: Heating Element and Control Device(s)

[0089] Various examples of heating element and control devices 2150 according to aspects of the present technology are described below. It should be appreciated that these examples are not intended to be exhaustive, and that features of the various examples may be excluded or included without departing from the present technology.2.1.1 Examples of Heating Element and Control Device(s)

[0090] In the example of FIG. 2D and FIG. 2E, an integrated heating element and control device 2150 includes a connection portion 2154 (e.g., made of an appropriate material for exposure to water, typically a corrosion resistant metal) providing a base for attachment to the water heater 2100. A resistive heating element 2102 extends from the connection portion 2154. A pressure and temperature sensor 2118 / 2122 is also received in connection portion 2154.

[0091] In examples in which the switching device 2112 is a solid state device, a heat sink may be provided to compensate for heat generated by operation which might otherwise degrade or destroy the device. In the illustrated example, a heat sink 2160 (e.g., a thermally conductive rod or heat pipe)protrudes directly into the water at the bottom of the water heater 2100 to reject the heat from the switching device 2112 directly into the water. In alternative embodiments, another type of heat sink (e.g., an air cooled heat sink) may be used.

[0092] FIG. 2F and FIG. 2G illustrate an exemplary switching unit 2113 in which the switching device 2112 and heat sink 2160 are provided as a separate unit to the resistive element 2102.

[0093] An electrical isolation portion 2156 is provided to the connection portion 2154, including recesses for locating the pressure and temperature sensor 2118 / 2122 and switching device 2112. The electrical isolation portion 2156 provides electrical insulation between the metal connection portion 2154 and a PCB 2157 providing connections to applicable components.

[0094] Remaining hardware may be provided in an electrical enclosure with wires connecting to external sensors. It is envisaged that the main electrical feed to the heating element 2102 may be routed through the electrical enclosure for power metering, and supply power to other low voltage components. This electrical enclosure may be located, for example, inside an enclosure of the heater element if there is room, or externally.

[0095] FIG. 2H illustrates an exemplary heating element and control device 2150 according to an aspect of the present technology. A water heater 2100 typically includes a water reservoir having a side wall 2108 defining an interior 2109. In this example, the heating element and control device 2150 includes a housing 2152 extending from connection portion 2154 into the interior 2109, in the same direction as resistive heating element 2102. It should be appreciated that other examples are contemplated, in which the housing 2152 and / or components are arranged in an alternative configuration. The connection portion 2154 may be configured to be fitted to a standard element port (e.g., screwed or bolted) of the water heater 2100 - for example to be retrofitted in place of a standard heater element.

[0096] Within the housing 2152 is provided temperature sensor 2118, and in this example vibration and / or acoustic sensing device 2120. Thermal and / or electrical isolation 2156 is provided within the housing 2152, to separate the sensing devices 2118 and 2120 from power control and monitoring devices (including the switching device 2112, thermal fuse 2158, and power meter 2116).2.1.2 Secondary Heating Source (e.g., PTC type heating element)

[0097] One aspect of the present technology utilizes a second heating source in addition to a traditional resistive heating element. In examples this second heating source may be a PTC type heating element. For example, with reference to FIG. 21, a PTC type element 2180 may be provided in addition to resistive heating element 2102. The PTC type element 2180 may have a lower power rating than the resistive heating element 2102 - for example, in the order of 80 W to 100 W for the PTC type element 2180, in comparison with a power rating in the order of 1 kW to 5 kW for the resistive heating element 2102. PTC type heating elements may be referred to as self-regulating heaters, having the capacity to maintain a particular temperature without use of a thermostat.

[0098] In examples, the PTC heating element 2180 may be installed integral to the resistive heatingelement 2102, immersed in the hot water tank, and / or connected to the exterior wall of the water heater (or a combination thereof using multiple PTC heating elements 2180).

[0099] In examples, the PTC type element 2180 may be configured to achieve a temperature below the upper temperature threshold of a thermostat controlling the resistive heating element 2102. In examples, the PTC type element 2180 may be configured to achieve a temperature below the lower temperature threshold of a thermostat controlling the resistive heating element 2102. In examples, the PTC type element 2180 may be configured to achieve a temperature below the upper temperature threshold, but above the lower temperature threshold, of a thermostat controlling the resistive heating element 2102.

[0100] It is envisaged that this configuration may be used to achieve a similar effect to that described further below, in terms of meeting passive load components of water heaters, but without requiring active control. For example, a lower power (e.g., 80 W) PTC element 2180 may be configured to heat up to about 65 °C and then maintain this level. The main resistive heating element 2102 may operate under typical thermostatic control (i.e., switch on and off based on upper and lower temperature thresholds), but the lower power PTC type heating element 2180 runs separately at a low rate until the heat hits about 65 °C. In the example of a household being away for a week, this approach would maintain or slowly increase the temperature - but without demand there would be no operation of the main thermostat. This changes a 3 to 4 times a day 3kW load to a continuous maximum load of approximately 100W. Even under normal operation it is anticipated that this approach may have a beneficial impact on the timing and number of thermostat operations.

[0101] It is envisaged that examples of the present technology may be utilized for purposes other than compensating for passive losses as described above. For example, most homes in North America are wired with a combination of 120V and 240V. The power demands of a conventional electric resistance water heater are such that connection to 120V outlets is impractical to meet peak load demands. Installation of higher capacity 240V wiring may present an impediment to the use of electric resistance water heaters. The present technology may be used to provide a continuous load at a lower power (e.g., 300 W) to an elevated temperature of 65+ °C using the PTC type element 2180, and then the main resistive heating element 2102 may be used to achieve the higher temperature threshold - but from a higher starting point.

[0102] FIG. 21 shows an exemplary heating element and control device 2150 enabling a continuous low energy heating operation independent from traditional thermostat control. It is envisaged that this load will be self-regulated (such as a Positive Temperature Coefficient - PTC - heater) or of a sufficiently low magnitude that continuous operation is unable to compromise the cylinder integrity and safety devices. The device 2150 has a connection portion 2154 (e.g., configured to be fitted to a standard element port - e.g., screwed or bolted - of the water heater 2100, allowing retrofitting in place of a standard heater element unit). The connection portion 2154 supports a main heating element 2102, anda secondary "trickle" element 2180 having a housing 2182 containing secondary heating device 2184 (e.g., PTC heater), and thermal cut out fuse 2158. In this example, the main heating element 2102 is controlled by thermostat 2110, and trickle element 2180 is self-regulating. While not illustrated, it is envisaged that in examples the respective heating elements may share a neutral wire, with separate live wires. Referring to FIG. 9C, the connection portion 2154 may be configured to provide a thermal break between the main heating element 2102 and trickle element 2180, for example using slot 2170.

[0103] FIG. 2J shows another exemplary heating element and control device 2150 having a similar configuration to that illustrated in FIG. 21 in terms of the main heating element 2102 being controlled by thermostat 2110, and trickle element 2180 being self-regulating. In this example, a temperature sensor 2118 is provided within housing 2152, integrated into the connection portion 2154. In this example, a pressure sensor 2122 is also provided in the connection portion 2154, exposed to interior 2109 of the water heater 2100 when installed. Referring to FIG. 2H, the connection portion 2154 may be configured to provide a thermal break between the heating elements (e.g., main heating element 2102 and trickle element 2180) and sensors (e.g., temperature sensor 2118 and pressure sensor 2122), for example using slot 2170.

[0104] FIG. 2K shows another exemplary heating element and control device 2150, configured to enable active control of main heating element 2102. In this example, the device 2150 includes an integrated temperature control unit 2110 and thermal cut out fuse 2158 for main heating element 2102, provided in connection portion 2154. In this example, a controllable switching device 2112 may be controlled by local controller 2200 to implement a control methodology for the main heating element 2102. The controllable switching device 2112 is configured to revert control to the integrated temperature control unit 2110 in the event of a loss of a control signal from the local controller 2200.2.1.3 Considerations for Retrofitting, Replacement, and New Installations

[0105] It is anticipated that the present technology may be implemented by way of retrofitting existing water heaters, and / or integration into a new water heater design (i.e., a water heater designed and dedicated to implementation of the present technology).

[0106] In examples in which the heating elements are replaced, it is envisaged that the integrated embodiments described above may assist with reducing the need for modification of the reservoir and / or piping.

[0107] In retrofitting examples in which existing hardware such as heating elements 2102 and thermostats 2110 are retained, it is envisaged that the various components described above may be installed separately. For example, a temperature sensor in the form of a thermocouple is placed in contact with the water heater wall proximal to the thermostat (e.g., the lower thermostat in case of multiple element water heaters). A pressure transmitter (potentially combined with a temperature sensor) may be installed, for example, in the following locations: on the inlet piping between the water heater and any potential check valve installed on the water supply, on the outlet water piping between the water heaterand the mixing valve (having the added benefit of recording the temperature at the top of the water heater), and / or installed on the water heater itself if there is a spare connection into the reservoir (again with the added benefit of additional temperature measurements). The remaining hardware may be packaged in an electrical enclosure with wires connecting the external sensors. The main electrical feed to the heating element may be routed via the electrical enclosure to enable measurement of characteristics such as current, and supply power to the low voltage components.

[0108] In this configuration it is contemplated that a solid state switch may or may not be included, with an appropriate heat sink.

[0109] In examples, the hardware described herein may be integrated into the design of a water heater for new installations. An exemplary installation may include a combined pressure and temperature sensor located proximal to a lower thermostat controlling an associated heating element. A thermal heatsink for a switching device (see, e.g., FIG. 2F) may be preferably located in close proximity to the heating element for wiring efficiency, but below the heating element to allow for heat rejection into the colder water at the bottom of the reservoir.

[0110] The reservoir may include dedicated ports or nozzles for installation of the various components exposed to, or projecting into, the interior of the reservoir.

[0111] It is envisaged that standardization of components such as piping, valves, pressure reduction and mixing valves integrated and insulated in an integrated water heater would assist with predictability for the purposes of control.3. System Insights

[0112] In the interests of ease of understanding, underlying principles of operation of the system are outlined below.3.1 Hot Water Heater Energy Model

[0113] A hot water heater can be considered in various states, e.g., either a homogenous volume at one bulk temperature, or as stratified temperature layers with a minimum of one distinct thermocline. Buoyancy forces dictate that the hot water is located above the cold water. Although the thermocline allows heat transfer across the surface, in the short term they provide a level of separation between the layers.

[0114] An abstraction of operation has been developed by the inventors based on theoretical and empirical observations for medium term operation, where the water heater is treated as two separate stores of energy: the hot water energy at the top of the water heater, and the cold water at the bottom of the water heater which is largely ignored.

[0115] The hot water at the top of the water heater is treated as a store of energy that is able to increase when the water in the reservoir is heated and energy makes its way to the top of the reservoir. The hot water energy at the top is consumed when the hot water outlet is opened, or through passive losses where heat is lost to the physical environment.

[0116] This model is a simplification that assumes the thermocline acts as a one way thermal barrier that allows the majority of additional heat energy to pass through from the cold layer (generated from the element heating cold water to hot water) to the hot layer, but does not let the energy to pass from the hot layer to the cold layer.3.2 Inferences of Flow from Pressure

[0117] Referring to FIG. 3, the pressure 3100 of the system embodied in a water heater is normally bounded by Pl and P2, where Pl is the supply pressure to the water heater (e.g., in an urban context, the town water supply downstream of the incoming pressure regulator), and P2 is the maximum water heater pressure. In examples, maximum pressure may be delimited by a high pressure expansion valve, or expansion vessel, fitted to the water heater (noting that this may not always be the case).

[0118] Excluding the influence of flow, the system moves between Pl and P2 by the addition of heat, or through passive losses. When a heating element is turned on, and heat is added to the water heater, the pressure increases from Pl to P2 because the volume is constrained (represented by first slope 3102). Conversely, when there is no heat being input into the system there are ongoing passive losses to the surrounding environment, which cause the temperature and pressure within the closed volume system to decrease from P2 back down to Pl (represented by second slope 3104).

[0119] Without considering the effects of flow, if the water heater is heated (first slope 3102) and the pressure reaches P2, it will remain at P2 (as represented by stable portion 3106) until the heat is turned off when passive losses begin to take effect (second slope 3004). While the system is represented by stable portion 3106, the high pressure expansion valve is operating, rejecting warm water to drain (increasing the energy losses from the system).

[0120] When the water heater is static and has cooled for an extended period of time the pressure in the system will sit at Pl. At this point the pressure will approximate the makeup water supply pressure (i.e., the pressure of water entering the system from the cold water supply). While the system is represented by Pl, then the contracting hot water volume (due to temperature reduction in a closed container) is made up by incoming cold makeup water, with cold water continually bleeding into the system and cooling the water heater shell and water within (thereby increasing energy losses from the system).

[0121] When the pressure drops below Pl (e.g., first pressure drop 3108 and second pressure drop 3110), flow out of the system can be inferred. During periods of outward flow, the pressure at the water heater will approximate the pressure losses in the downstream system, from the water heater to the point of use (e.g., a shower or tap). The higher the flow rate (shower or full open tap) then the higher the system losses, which corresponds to a lower pressure reading at the water heater (e.g., as shown by first pressure drop 3108).

[0122] Similarly, if the point of use is a low total flow then the system losses between the water heater and the point of use will be lower. As a result, the pressure reading at the water heater will be relativelyhigher (although still less than Pl) than the reading for a high flow pressure drop 3110.

[0123] Using this methodology, a differentiation may be made between high and low flows. This may be used to inform determination as to the energy state of the water heater (e.g., calculation of state of charge - as described below) and subsequent control.

[0124] Operating the water heater at either Pl or P2 increases energy losses from the system, either by rejecting hot water that will eventually be replaced by cold water, or by directly drawing in cold water. Therefore, it is believed operating for longer periods of time between these identified layers may be advantageous to reducing energy losses. As will be described further below, control methodologies of the present technology may utilise pressure signals as an input to controlling heating to extend the system operation time between Pl and P2 (e.g., after long periods of static cooling, or directly after a flow event). It is anticipated that this is likely to involve more frequent and shorter heat addition on average.

[0125] It is believed that pressure may be a useful means for inference of flow, at least in part because available pressure sensors have a relatively high degree of reliability. However, it is also envisaged that similar data may be gathered from alternative indicators of flow, such as temperature sensors (e.g., in a string, or point source locations), flow meters or switches, or acoustic devices to detect flow.3.3 Thermal Quality of Flow

[0126] The supply of heated water to the user, downstream of the mixing valve, is a combination of hot water from the water heater tempered with cold water from the main supply. This final supply temperature or demand temperature (typically in the order of 50-55°C) sets a reference energy state of the water heater (referred to herein as SOC_datum).

[0127] The inventors have developed a metric, Heat Quality Factor (referred to herein as "Quality"), to capture the value of storing water at a temperature above the required user temperature (i.e., the temperature at the outlet of the mixing valve).

[0128] If the hot water exiting the water heater and entering the mixing valve is above the mixing valve temperature, then it has a higher "quality" than hot water entering the mixing valve at the same temperature as exiting the mixing valve (i.e., because it can mix with more cold water while still achieving the desired output temperature).

[0129] When a hotter (i.e., "higher quality") water enters the mixing valve, it is tempered with more cold water. Both the hot water and the cold water entering the mixing valve contain enthalpy, or internal energy. Therefore, even the cold water entering the mixing valve contributes energy to the final mixed water temperature, reducing the amount of energy that needs to be withdrawn from the water heater. For example, referring to FIG. 2B, where the stored water temperature is 50°C and the demand temperature is 50°C, the water heater is entirely relied on to provide the required energy (i.e., no energy is supplied from the cold water). In contrast, referring to FIG. 2C, where the stored water temperature is 70°C and the cold water temperature is 15°C, a portion of the energy in the mixed water at the demand temperature of 50°C is supplied by the cold water (thereby reducing demand from the water heater).

[0130] As will be described further below, an initial value of quality may be calculated based on the Calibrated SOC. The quality is proportional to the ratio of the Calibrated SOC to the SOC at the mixing valve outlet temperature (SOC_datum).

[0131] In alternative examples, quality may be sensed using a temperature sensor in the hot water outlet - however the methodology of estimating the quality based on Calibrated SOC reduces the complexity and costs associated with this additional sensor.3.4 Energy Components

[0132] A simplified energy balance for a water heater is shown below:where (over a period of time): Q_heat is a times sum of heat energy input; Q_flow is p x sum of the duration of flow events divided by the quality; and QJosses is y times duration times quality.

[0133] The coefficients (a, , and y) are derived, for example, using machine learning and / or regression approaches.

[0134] QJieat: the energy associated with heating is obtained, for example, through continuous monitoring of the power flow into the heating source.

[0135] QJosses: the energy associated with passive losses is typically well bounded within a range by physical constraints. The primary drivers of passive losses may include: the ambient physical environment where the heater is located; the quality of the insulation of the heater and associated connections; and the difference between the heater internal temperature and the ambient temperature.

[0136] Typically, the range of the passive losses of the water heater and close connected piping (i.e., not the entire connected system) may be assumed to be in the range of 2 to 6kWh / day depending on the quality of the insulation and the temperature. This equates to an average energy load of between 80W (i.e., 0.08 kW) and 250W (i.e., 0.25 kW).

[0137] Q_flow: the energy associated with flow has more potential variability than passive losses, but is also bounded within a range by physical constraints. The primary drivers of this variability may include: variations in the physical flow rate (noting this is generally limited to the capacity of the system); types of flow (e.g., short intermittent flows in comparison with longer continuous flows); and the quality of heat of the water entering the mixing valve.3.5 Energy Decomposition

[0138] Improving the accuracy of determination of the flow and energy losses from the water heater involves the application of algorithmic approaches to periodic windows between subsequent calibrated SOC values (described below). Cumulative heat energy input over the window is a known input, which leaves the energy flow out of the water heater as the unknown parameter to be calculated. The energy flow out of the water heater is broken into passive losses and flow losses which can be determinedthrough regression or other machine learning (ML) approaches.

[0139] In an operating system, the rate of energy lost during a flow event may be significantly greater than passive losses (e.g., in the order of 70 times greater), emphasising the importance of flow to accurate estimation of the state of charge.4. Environment State System

[0140] While examples provided herein may describe implementation of various calculations within the cloud-based environment state system 3000, it should be appreciated that some or all of such calculations may be made within the local environment 2000. For example, determination of the state of charge of a water heater may be performed by timeseries database 3004, or by local controller 2200.

[0141] Further, examples are contemplated in which the output(s) from the environment state system 3000 are delivered to external products or services - for example via system API 3500. For example, the information may be utilized by a service co-ordinating control and optimizing of disparate energy devices.4.1 State of Charge

[0142] At a high level, the State of Charge ("SOC") of a water heater may be determined using various conditions of the hot water heater. As previously noted, the SOC of a water heater is a representation of the current energy stored relative to the full capacity of the device.4.1.1 Continuous Estimation of State of Charge

[0143] FIG. 4A illustrates a method 4100 of continuously estimating SOC of a water heater. In a modified State of Charge ("mSOC") calculation step 4102, an initial estimation of the State of Charge of a water heater is made based on the observed temperature S004. Referring to FIG. 4B, the illustrated chart represents an algorithm mapping the observed temperature to mSOC - e.g., multiplying the initially calculated SOC by a modifier function to generate mSOC.

[0144] For completeness, it should be appreciated that the example of FIG. 4B is intended to demonstrate the approach of augmenting the limited available information (i.e., the real temperature measurement) rather than stipulate a strict adherence to the illustrated function.

[0145] This temperature based mapping of the stored energy is adjusted to reflect both the available useful stored energy, and the confidence in the resulting calculation as a calibrated SOC.

[0146] Returning to FIG. 4A, if this initial calculation meets a threshold high confidence level it provides a calibrated SOC 4108 to start an ongoing estimation window 4110. If the initial calculation does not meet the threshold, heating 4112 is initiated until the threshold is achieved.

[0147] Incremental time steps 4114 are undertaken and energy changes are calculated for discrete energy components: flow 4120, passive losses 4130, and heating 4140.

[0148] Calculation of Q_flow 4122 is based on flow detection 4116, Q_flow coefficient 4124 (derived, for example, using machine learning and / or regression approaches 4150), and Quality 4160. Calculation of QJoss 4132 is continuous and based on QJoss coefficient 4134 (derived, for example, using machine learning and / or regression approaches 4150), and Quality 4160. Calculation of Q_heat 4142 is based onheating 4118, and power meter output 4144.

[0149] Q_flow, QJoss, Q_heat are summed in summing step 4170, and used to generate an estimated State of Charge ("eSOC") in step 4172.

[0150] The eSOC is continually updated and periodically compared to the current mSOC in comparison step 4174 to determine if the calculated value better represents the SOC. If so, the mSOC becomes the Calibrated SOC 4108 in update step 4176.

[0151] Cumulative energy / time data 4178 is transmitted to and maintained in time series database 3004.

[0152] In jurisdictions such as New Zealand, where building code compliance law requires stored hot water be heated to a minimum temperature (e.g., not less than 60°C) at least once during a fixed period to maintain legionnaires control, there is a sanitising timer 4180 and record of when the temperature was at a sanitizing temperature.4.2 Ancillary Data

[0153] The ancillary data 5000 may be obtained from a variety of sources, for example databases capable of being accessed via APIs. This data may be used by the adaptive control module 4000 to adjust operational strategies, whether from the perspective of the account holder / consumer associated with a given local environment 2000, an operator of the power network (e.g. a utility company), and aggregator, or any other stakeholder.

[0154] In examples, the ancillary data 5000 may include one or more of: Weather data (e.g., real-time data on weather conditions such as temperature, humidity, precipitation which could affect consumption patterns); Energy pricing data; Grid Demand and Constraints (e.g., information on peak demand times, and potential grid constraints to enable load shifting); Non-Dispatchable Data (e.g., data on the current and predicted availability of renewable energy sources to enable alignment with sustainable energy consumption); Frequency Data (e.g., in additional to regional grid data, this data may include highly localized and / or fast changing data); Voltage Data (e.g., in additional to regional grid data, this data may include highly localized and / or fast changing data); Geographic Cold Water Temperature; Geographic Maintenance Data; Regulatory Constraints (e.g., any operational constraints imposed by energy regulation policies - for example a policy that a certain percentage of energy must be regionally or locally generated); Local Holiday Data; Short-term Trend Data (e.g., general trends over shorter time periods - i.e., daily and / or weekly); Seasonal Trend Data (e.g., general trends over seasons); Curtailment; and / or Carbon emission intensity.

[0155] FIG. 5A illustrates the variation over time in factors that may influence the supply of electrical energy, and may therefore be relevant as ancillary data in control strategies for water heaters in accordance with aspects of the present technology. It may be seen that the ancillary energy data 5000 may include data that is monitored in real-time (such as voltage and / or frequency), along with future conditions (including both scheduled and projected conditions).4.2.1 Responsive Load Adjustment for Non-dispatchable Variability

[0156] Certain renewable energy sources such as wind, and especially photovoltaic (PV), are referred to as non-dispatchable energy sources (i.e., there is limited ability to generate power on demand due to the nature of the energy source). As shown in FIG. 5B, such non-dispatchable energy sources often have a high degree of variability in terms of power generation over time. For example, wind generation 5200 may occur through an entire 24-hour period, but is subject to a range of variable factors such as wind speed and air density which can lead to a high degree of intermittency. As another example, even ideal PV generation 5202 is delimited to daylight hours, but real-world PV generation 5204 is further influenced by shading due to cloud cover.

[0157] As will be described further below, aspects of the present technology may use both monitored and predicted characteristics of non-dispatchable energy sources in decision making and control. The present technology enables a correlation of load with non-dispatchable generation sources to reduce curtailment of load, and therefore lessen the chance of continuing load during a temporary reduction in generation. Such peaking power loads may otherwise require the use of dispatchable power systems (known as peaker power plants) to meet demands.

[0158] As a non-limiting example: a community has a high penetration of solar PV installations. During a sunny day there is an excess of solar energy generation, but cloud cover can quickly reduce generation capacity. If the demand remains high, then at short notice additional peaker power plants may need to be deployed to meet demand. Alternatively, if the demand is low then the excess solar energy will be sold to the grid at low compensation rates, or may need generation to be curtailed.

[0159] While actual implementation may account for multiple factors, for ease of understanding the control implemented according to aspects of the present technology may increase heating during times of excess solar generation and pause during periods of reduced generation (e.g., when clouds pass). This not only utilizes the solar energy efficiently, but also avoids the need for peaking power plants to compensate for the sudden drop in solar output.

[0160] Similar considerations apply to wind powered generation. If a wind farm experiences a sudden drop in wind speed, the electricity generation could decrease unexpectedly. In such cases, the control system of the present technology could delay the supply of electricity to decrease the power demand, thus maintaining grid stability without resorting to non-renewable backup systems.4.3 Data Processing and Profile Generation

[0161] The various measures described above can be used to represent the stored energy within the system of a water heater.

[0162] The primary metric is the current estimated SOC of the system (e.g., expressed in kWh). This represents the energy in the system that is believed to be available at the current point in time.

[0163] Other metrics may be useful in providing measures of flexibility, in terms of the ability to both delay consumption of energy, and also indicate a measure of the urgency for heating. These measuresmay typically be dominated by a combination of two or more of: Element Heating Capacity (i.e., recovery capacity), Volume of reservoir, Maximum Stored Temperature, Expected Demand Profile, Heat Loss Rate, Current Water Quality, and Current estimated SOC.

[0164] Most residential users will follow a typical demand pattern. In examples, user behaviour prediction module 3008 may implement Artificial Intelligence (Al) and Machine Learning (ML) techniques for pattern recognition. By analysing collected context information (e.g., from timeseries database 3004) user behaviour may be recognised and predicted, for example in order to forecast near-term demand based on historical usage. The user behaviour prediction module 3008 may further detect deviations from normal patterns through context signals, and adjust predictions to account for anomalous behaviour.

[0165] For demonstration purposes, the graphs shown in FIG. 6A to FIG. 6C indicate aspects of an exemplary approach to establishing an expected relative stored energy metric. This approach reflects a discrete simplified version of energy space measurement.

[0166] In examples, a service buffer is established over a period of time. The service buffer is a value that dynamically reflects the user's expected profile to establish a region where the quality of service (i.e., maintenance of hot water supply) is unlikely to be compromised. This curve is primarily determined by the relationship between the user consumption profile and the water heater capacity. A small heater requires maintenance of a higher relative state of charge to maintain service than a larger heater.

[0167] The predicted load profiles of several statistical load profiles are projected from the current known state. These profiles are weighted for the current quality of the hot water heater outflow and incorporate passive losses.

[0168] The relative stored energy is the net area between the profile and the service buffer profile. It provides a measure of urgency for additional heating, or ability to delay heating, beyond the simplicity of the base SOC metric.

[0169] FIG. 6A shows a water heater that has a small capacity relative to the expected user demand. The short term forecast is small but positive, and the long term forecast is overall negative - as can be seen by the greater area under the minimum service buffer line than above the line.

[0170] The impact of a similar user demand profile in a 30% larger capacity heater is shown in FIG. 6B. The initial state is higher, and the service buffer profile is lower, providing a significantly increased relative stored energy. FIG. 6C shows the profile of a larger capacity tank having a greater daily demand.

[0171] The value of the relative stored energy is considered important in increasing operational flexibility, and enabling geographical based control. The ability to measure this metric enables tuning of the necessity for heating to more optimal times with reduced loss of service.

[0172] It is also envisaged that this metric may enable prioritisation of load relative to an individual heater among multiple devices operating in a constrained manner within a Virtual Power Plant (VPP).4.4 Digital Twin

[0173] According to aspects of the present technology, a digital twin may be established for individualwater heaters using the data collected and calculated as described herein.

[0174] A digital twin is a digital model of an actual real-world physical product, system, or process that serves as the effectively indistinguishable digital counterpart, the state of which is updated in effectively real-time using the collected data.

[0175] It is envisaged that data input to the digital twin may include, for example: Location (e.g., GPS location and / or other indication of location within the nodes of an electrical network), Energy Rates (e.g., Q_heat, QJoss, Q_flow), Current SOC and Maximum SOC (i.e., SOC at maximum temperature), Quality, Heating status (i.e., is power being drawn); Flow status, Time since Sanitising Temperature, potential Predictive Load Paths, the amount of stored energy within the water heater relative to its projected demand , and the total capacity of the water heater and its percentage charge.5. Adaptive Control

[0176] In examples, the adaptive control module 4000 algorithm provides the basis of an energy management system, configured to enhance operational efficiency, and adapt to user patterns and / or system demands. It will be appreciated that functions are described herein within a holistic context such that both individual consumer requirements and network demands are accounted for. However, this is not intended to exclude examples in which interests of one party is prioritized over the other. For example, in some embodiments the interests of the individual consumer may be prioritized in terms of reducing energy costs and ensuring availability of hot water, and / or operate entirely independently of other consumers. In alternative embodiments, the network's interests may be prioritized over those of the individual consumer.

[0177] In examples, the adaptive control module 4000 algorithm takes the environment state data and predictive information, and leverages a multi-faceted approach as detailed herein. In examples, the adaptive control algorithm includes constraint generator components which: incorporates a neural network modeler to interpret complex data relationships and forecast system needs; and utilizes inequality and equality constraint generators to outline the permissible operational parameters, ensuring the system adheres to physical and regulatory limitations.

[0178] In examples, the adaptive control algorithm further includes an integrated optimization method that: receives and processes operational data from individual energy units, including geographical location within the electrical network, performance metrics, demand profiles, and operating conditions; integrates contextual information such as location-specific data, grid constraints, and renewable generation variability; employs various optimization algorithms, including reinforcement learning and linear optimization, to fine-tune energy system operations; responds to real-time changes in the energy landscape, adjusting operational strategies to optimize efficiency, cost, and environmental impact; continually updates the model with new data, potentially switching algorithms as conditions evolve, thus refining decision-making processes over time.

[0179] The Al implemented control algorithm orchestrates a comprehensive and responsive system.It adjusts to the fluctuating landscape of energy needs and generation, particularly focusing on integrating renewable energy sources effectively. This system is not just reactive but also anticipatory, aiming to deliver optimized energy solutions that are both user-centric and grid-responsive.5.1 Individual Operation as a Distributed Energy Resource (DER)

[0180] In examples, a reinforcement learning / algorithmic control strategy may be implemented that allows the system to interact with the world on a 'greedy' basis - i.e., optimized for the maximum benefit of an individual enabled hot water heater acting as a stand-alone distributed energy resource (DER).

[0181] This control strategy may utilise the output from the real time digital twin as described above, particularly the state of charge (or stored energy) of the water heater, the amount of energy consumed in a defined period, and the predicted forward load demand from the water heater.

[0182] It is envisaged this information may be combined with external API signals including, but not limited to: the amount of curtailed energy (current and projected); the percentage of renewable energy generation (current and projected); the price of power (current and projected); network time of use tariffs (current and projected); the amount of carbon emissions (current and projected); and any network constraints (e.g., predicted high capacity events or maintenance events) - whether current or projected.

[0183] In examples, the control algorithm may be biased towards a desired outcome - for example, reducing consumer cost, or carbon emissions.5.2 Aggregated Control

[0184] An increasing number of distributed energy resources have been introduced to the grid, able to quickly respond to network signals. However, this capability has led to new challenges, such as localised or general secondary peaks and herding behaviour, which occur when many customers compete to use power simultaneously.

[0185] A water heater under energy state based digital twin control has significantly higher flexibility potential than traditional thermostat control. In examples, a plurality of water heaters may be operated in aggregate as a Virtual Power Plant (VPP). It is considered that the granularity of data regarding the energy state of individual water heaters (enabled by the present technology and embodied in digital twins), and associated control of those heaters, enables a cooperative approach to load sharing that cannot be achieved with conventional technology.

[0186] The aggregation of water heaters into a VPP may be determined based on various factors, such as electricity supplier, or more commonly geographic connection within the electrical network.

[0187] Referring to FIG. 7, the system 7000 may comprise a real-world environment 7100 comprising a plurality of water heaters 2100 having a geographic commonality in terms of connection within a network (e.g., supplied via a common station transformer 7110). The plurality of water heaters 2100 may also belong to lower-level groups - for example a first local transformer 7112A supplying first subgroup 7114A, and a second local transformer 7112B supplying second subgroup 7114B.

[0188] A digital twin environment 7200 mirrors real-world environment 7100 - including digital twins7202 selective grouped in one or more VPPs (for example, substation level VPP 7220, first local VPP 1 2.T. , and second local VPP 7222B). Sensor data is transmitted from the water heaters 2100 to Real Time State Generation module 7300. The output from Real Time State Generation module 7300 is input to Historical Flow / Load Data module 7302, which is in turn used by real Time Flow / Load Prediction module 7304 to output to a Cooperative Adaptive Control Algorithm module 7306. Outputs from the various modules also feed into the digital twin environment 7200, which feeds state information to Cooperative Adaptive Control Algorithm module 7306.

[0189] Ancillary data 7400A / 7400B / 7400C specific to the respective VPPs may also feed into the cooperative adaptive control algorithm 7306.

[0190] In examples, the Cooperative Adaptive Control Algorithm module 7306 may operate a cooperative multi-agent reinforcement learning (MARL) algorithm, in which multiple agents work together to achieve a common goal in a shared environment. In the present technology, the agents are water heaters acting as loads in distributed Virtual Power Plants.

[0191] The energy state information provides both an individual current energy state, but additionally the ability to compare and prioritise individuals within an aggregate (e.g., House A has lower current stored energy, but low overall expected demand in comparison to neighbouring House B which has higher current stored energy and a higher expected demand).

[0192] Referring to FIG. 8A, traditional thermostat control may be inefficient with respect to use of renewable or non-dispatchable energy. In examples in which algorithmic control of each water heater prioritises the same outcome individually (e.g., controls based on lowest price), herding behaviour may result in peak demands being created which negate the benefits of the control - as depicted in FIG. 8B.

[0193] In contrast, the aggregated control methodology of the present technology may implement individual control to achieve collective objectives. This may enable load shaping (as shown in FIG. 8C) that optimises use of desirable energy (such as from non-dispatchable sources) across the day.

[0194] As a first example, local PV generation may follow a generation profile over the day. A collaborative algorithm can signal relative availability, and try to manage the net load of many agents cooperatively to match the profile and minimise curtailment.

[0195] As a further example, load may need to be reduced at a geographic location due to some network constraint. It is possible to manage this constraint in a cooperative manner to prioritise constraint across the collective of the VPP to minimise the loss of service to individuals. This may involve rapid switching between individuals in response to energy state to meet some VPP constraint (e.g., a maximum peak demand).

[0196] In examples, the algorithm may operate in a hierarchical manner - prioritising individual control (e.g., to reduce cost, or achieve low carbon footprint) in the absence of a higher priority need for a cooperative approach.

[0197] In examples, the sensors of the present technology may feed back local voltage levels tonetwork operators. In examples, the real-world environment 7100 may include local generation means 7120 (e.g., PV generation). In examples voltage at the locality (e.g., from the digital twins) may be mapped to local PV availability, providing another potential control signal.5.2.1 Mitigation of Load Management Issues

[0198] In examples, the present technology may be used to mitigate challenges associated with load pickup following load management events through aggregated control of one or more virtual power plants (VPP) composed of networked hot water heaters.

[0199] FIG. 9 illustrates how load pick up and / or rebound can affect network grids under bulk load control methodologies. The real time demand load is represented by the solid line. At the time when the projected load, represented by the dashed line, is expected to get high the network dispatches a demand response signal (labelled "on"). Upon receipt of this signal Distributed Energy Resources (DER) turn off, which drops the load (having the net effect of dropping the real observed load below the projected peak). However, when the demand response signal is turned off (labelled "off") there can be a load pick up where all of the DERs all turn on at once. This load pick up and rebound spike can, for a short time, be higher than the original projected peak demand if the integration of DERs back into the grid is not managed well.

[0200] To alleviate this, the aggregated control of the present technology leverages real-time knowledge of the energy state of each water heater within the VPP, and integrates this data with broader grid signals and actual consumer demand to implement a prioritised control strategy.

[0201] The present technology provides the ability to proactively distribute a constrained, low- capacity load across multiple water heaters based on priority and / or need, rather than waiting for the end of a load management event and risking a synchronised reactivation of DERs. The VPP continuously maintains a minimal, widely distributed heating load. This approach targets users based on their water heaters' current state and predicted near-term demand, ensuring that immediate hot water needs are met while preventing any significant surge in overall demand. This smooths out potential demand spikes to support grid stability. The VPP continuously adjusts its distribution of this low-capacity load based on real-time grid conditions and changing user needs, maintaining a balance between meeting hot water demands and optimising grid performance.6. User Analytics and Reporting

[0202] The rich data collected through operation of the present technology may be used to obtain insights regarding a variety of interest points.6.1 Real-Time Flow Data and Customisable Alerts

[0203] The incorporation of real-time flow data may significantly enhance the system's effectiveness and user experience. By continuously monitoring hot water flow, the system gains precise insights into actual usage patterns, enabling more accurate predictions of hot water demand. This real-time data is not only valuable for system optimization but also provides direct benefits to users through customizable alerts.

[0204] In examples, users may set personalised alert thresholds based on their specific needs and preferences. For instance, a user might set an alert for five minutes of continuous hot water consumption. This proactive notification empowers users to make informed decisions about their water consumption, potentially adjusting their usage. Moreover, this feature enables users to actively participate in energy management, fostering a sense of control and engagement. By bridging the gap between system-level optimization and individual user needs, real-time flow data and customisable alerts represent a significant advancement in hot water management technology.6.2 Comparative Consumption Reporting

[0205] According to examples of the present technology, the system may generate comparative consumption reports that contextualize a user's hot water and energy usage. For example, the system may compare consumption to similar households, encouraging energy conservation through social comparison.

[0206] In examples, these reports may include metrics such as average daily consumption, and efficiency rankings. In examples, these reports may include guidance on reducing power consumption (e.g., energy saving tips), personalized to the user based on consumption habits.

[0207] It is envisaged that this feature may assist not only with motivating users to make more conscious decisions about their usage, but also provide valuable data for utility companies and / or policy makers to design targeted energy efficiency programs. By integrating comparative reporting with realtime monitoring and alerts, the Virtual Power Plant (VPP) of hot water heaters becomes a comprehensive platform for promoting energy conservation at both individual and community levels, aligning user behaviour with broader sustainability goals.6.3 Determination of Passive Losses and Heater Efficiency

[0208] The system's comprehensive monitoring capabilities enable accurate measurement of passive heat losses in individual water heaters, providing valuable data for users and operators. By tracking temperature changes, energy inputs, and water usage, the system can identify underperforming units, enabling proactive maintenance or replacement decisions.

[0209] In examples, this information is presented to users in energy usage reports, empowering them to make informed decisions about heater upgrades.

[0210] On a broader scale, this data allows for the identification of consistently underperforming heater models, informing procurement decisions and energy efficiency standards.

[0211] In examples, a Virtual Power Plant (VPP) may optimise heating schedules based on this granular performance data, prioritising water heaters with higher passive losses to minimise overall system energy waste. This capability allows the VPP to act as a proactive efficiency optimisation platform, aligning individual heater performance with system-wide energy conservation goals.7. Passive Losses in Water Heaters

[0212] Water heaters often undergo significant periods of time without flow demand. During this 1time, passive cooling and then heating cycles occur under traditional thermostatic control. The heating operation causes pressure increases within the tank due to the expansion caused by the increased temperature. To avoid an overpressure situation without an expansion tank requires an expansion valve.

[0213] In essence, the reservoir of the water heater heats sufficiently to cause the expansion valve to operate and a typically small amount of warm water is rejected from the reservoir to waste, resulting in some energy loss. For example, over a period of 24 hours with no demand usage, thermostatic operation heating may occur at three points (6 am, 4pm, and before Midnight), representing the heating cycles triggered when water temperature drops below the thermostat setpoint. With no usage demand the heating energy addition is solely due to compensating for passive system energy losses. The passive losses occur continuously but are only compensated for during these heating cycles, resulting in an arbitrary timed load with little operational control of this load. Referring to FIG. 10A, an exemplary demand scenario with conventional thermostatic control is shown, in which heating occurs at three points (approximately 3 am, Noon, and 10 pm), representing the heating cycles triggered when water temperature drops below the thermostat setpoint. These cycles are caused by the combination of passive system losses and user demand. These three heating periods are due to the water heater heating to recover system energy (typically controlled on operating temperature). The heating load requires compensating for passive losses as well as the energy used for heating up water due to demand. The passive losses occur continuously and in this example are compensated for during these heating cycles, which is typical operation with regular user demand, resulting in the concentration of the recovery of the passive losses typically with the user demand energy recovery.

[0214] While there is flow demand cycling these losses are somewhat inevitable, but during long stretches (e.g., overnight or while unused but still powered on), it has been identified that there is an opportunity to mitigate these losses by adding heat energy at a rate approximately proportional to the passive losses and maintaining a relatively static state of charge.

[0215] Aspects of the present technology may treat the demand profile of a water heater as including a base load (including the continuous passive losses), and an active load (driven by consumer demand). As will be discussed further below, the present technology enables shifting of a portion of the overall load from the more volatile active load to more of a base load.8. Higher Frequency Incremental Heating

[0216] According to one aspect of the present technology, a control methodology of a water heater may be implemented in which frequent, incremental heating is performed in order to approximately match the rate of passive heat loss within the system. By monitoring and countering passive losses with corresponding heat input, a portion of the hot water system's load may be transformed from an uncontrolled, cyclical pattern to a managed baseline component, complemented by less frequent and lower amplitude peaks. For instance, instead of the large demand spikes associated with typical three- cycle thermostat operations — often comprising drastic shifts such as 2kWh, 6kWh, 2kWh — the methodutilises consistent, smaller heat additions. These additions may be calculated to closely align with the passive heat loss, which can account for up to 30% of the total load. As a result, the system may be enabled to maintain a more stable thermal state, reduce the frequency and magnitude of peak heating events, and transition from sporadic, uncontrolled energy draws to a more predictable and steady consumption profile. Ultimately, this method may offer a dual benefit: enhancing the efficiency of the hot water system while concurrently easing the demand on the electrical grid, especially during peak periods.

[0217] Referring to FIG. 10B, a control methodology is implemented in which heat is added incrementally throughout the day at a rate equivalent to the passive heat losses. Additional energy supplied to meet active demand is shown, and in comparison with the traditional control methodology of FIG. 6B it will be observed that the peak demands are noticeably shorter, indicating a reduction in the size and frequency of demand spikes. This continuous, baseline heating converts what is typically an uncontrolled load into a managed load with a predictable base component and infrequent, less intense peaks. By maintaining the temperature more consistently, this approach avoids the substantial temperature drops that trigger large energy draws in conventional systems.

[0218] It is envisaged that a PTC heating element 2180, such as that of exemplary heating element and control device 2150 shown in FIG. 21, may be sized to continuously heat at a rate close to the passive losses. This PTC element is thermally self-regulating (i.e., has a maximum temperature at which it will stop heating) and may be powered in parallel to the main heating element.9. Component Based Scheduled Load Distribution

[0219] One aspect of the present technology allows for management of individual units (i.e., local environments 2000) as part of a wider system, for example in order to improve use of low-cost renewable energy within network capacity constraints. A theoretical example is set out below, using a sample size of 100 water heaters all under control in the same geographic location. The basis for control assumes all of the water heaters are functionally identical, having a 4 kW heating element, and a demand of 12 kWh per day (including 3 kWh of passive losses, and 9 kWh of demand). This requires 3 hrs of heating per day (3 hrs x 4 kW = 12 kWh) per water heater. In this example the water heaters follow a typical domestic load profile and behave in a similar manner with morning and evening peaks. It will be appreciated that actual implementation would account for variability in water heater size, usage patterns, and loss rates.

[0220] Under traditional thermostatic control, for the total system heating load per day there is 1,200 kWh (1.2 MWh) of energy used, of which 300 kWh are passive losses, and 900 kWh relate to hot water demand. During morning or evening peak demand operation it is likely that approximately 30% of the water heaters may be on concurrently, resulting in peak system loads of approximately 120 kW (i.e., 30% of 100 x 4 kW). Of this load some portion is replacing passive losses that occurred in the hours prior to user demand.

[0221] In order to improve control of the load, the present technology aims to maintain an approximately constant state of charge when there is no active heating and during periods of no userdemand. To achieve this, the present technology operates the water heaters for short bursts of operation at regular intervals (or by a supplemental smaller element operating continuously) proportional to the passive load losses, spreading the load throughout the day. In the example above this is 300 kWh in passive losses, which is an approximately continuous system load of 12.5kW to replace the passive losses of the system. To achieve this, each water heater (of the total 100 water heaters) would need a continuous heat input of approximately 0.125 kW. In the case of burst operation, each water heater could be turned on for approximately 112 seconds every hour. This could be achieved for example by heating each cylinder for approximately 38 seconds every 20 minutes. With 100 water heaters under control, to distribute this base load of passive losses at any one time there will be approximately 3 water heaters turned on. The net result is that although the demand portion of the load is unchanged there is increased controllability in that portion of this load as the portion of the load that would have been attributed to the passive load is reduced.

[0222] The remaining demand load can then be dispatched at times when, for example: The individual local environments 2000 require it, including in scenarios in which the demand is predicted in advance of when it is needed; The power is generated by volatile renewables; The price of power is inexpensive (proportional to high concentration of renewable energy); To avoid time-of-use tariffs or other peak tariffs; To avoid peak times - noting that because the demand load is spread out, or targeted towards renewable energy, the traditional peaks are flattened.

[0223] The above operation is visualized in FIG. 11A and 11B, in which the number of cylinders is reduced to 4 (labelled Ato D) for ease of understanding. In FIG. 11A the water heater operation (separated into demand load and passive losses) is stacked to represent a typical residential demand profile. In FIG. 11B, the same load is controlled such that: The load due to passive loss is spread throughout the day; The load due to demand is incentivized to absorb as much non-dispatchable energy (in particular PV energy) as possible; and Some of the demand load is consumed overnight in preparation for use in the morning (i.e., before it is required).

[0224] As a result of controlling the load in this manner, peak loads are reduced and the overall demand is flattened.

[0225] FIG. 11C illustrates an alternative control arrangement in which the passive load component of the water heaters is met by more frequent operation of the water heater elements at lower power. For example, using rapid switching, the heating elements may be turned on for much faster periods (e.g., several phases), effectively converting a heating element into a lower power equivalent (e.g., reducing the effective output from 4 kW to 200 W).10. "Trickle" Heating Control

[0226] FIG. 12 shows another exemplary local environment 2000, including a second type of heating element 2180 for providing the small continuous heating according to the control approaches described above (which may be referred to herein as "trickle" heating for ease of reference). One or more pressuresensors 2122 are also provided. For completeness, it is envisaged that exemplary heating element and control devices described herein may have utility irrespective of whether the adaptive control methodologies of the present technology are implemented.

[0227] It is envisaged that the use of continuous trickle heating, potentially with state of charge monitoring and control, may provide for significant additional utility in locations or situations where power supply is constrained - e.g., operating on a 120 V supply voltage. The present technology may enable this constrained line to supply more effective operation of a water heater without the necessity to upgrade local wiring.11. Supplementary Energy Storage for Lower Mains Voltage

[0228] The present technology may be adapted for use in jurisdictions specifically having low capacity installed wiring (e.g., in the order of 110 to 120V). Most homes in North America are wired with a combination of 120V and 240V. The power demands of a conventional electric resistance water heater are such that connection to 120V outlets is impractical to meet peak loads, and installation of higher capacity 240V wiring may present an impediment to the use of electric resistance water heaters. It is the inventors' understanding that 120V water heater offerings in such markets utilise heat pump technology, which are expensive and have limitations around efficient operation based on average ambient air temperature.

[0229] However, a 15A receptacle outputs about 1.44kW based on typical circuit design constraints (although in practice, as a continuous load this would encounter some reduction in maximum load). In a day, this approximates to 28.8kWh of available energy, which is above an expected mean user demand of ~12-15kWh. In other words, a standard 15A / 120V outlet is capable of meeting the average daily power demands of an electric resistance water heater, but would be unable to supply peak load constraints for practical operation.

[0230] In aspects of the present technology, systems intended for implementation in jurisdictions with a lower nominal mains voltage (e.g., 120V in North America) may include a supplementary energy storage device such as a battery. In examples the battery (including any applicable housing and associated components) may be integrated into the water heater as a unit.

[0231] FIG. 13 illustrates an exemplary electric water heater 2100 configured to be connected to a lower voltage (e.g., 120V) AC power supply 14000. A supplementary energy storage device in the form of battery 14010 is connected to the AC power supply 14000, and supplies power to an upper heating control device 14020 and associated DC heating element 14022. A lower heating control device 14030 is also connected to AC power supply 14000, and controls AC heating element 14032.

[0232] In examples, the upper heating control device 14020 may be configured to operate a thermostatic control, set to heat DC heating element 14022 on reaching a nominal temperature value (e.g., 60°C). For example, a 3kWh battery 14010 may be capable of delivering power in the order of 3- 9kW. It is envisaged that such a battery capacity in combination with a ~4.5kW DC heating element 14022may be sufficient for peak residential loads (noting that heating via DC heating element 14022 is supplementary to AC heating element 14032).

[0233] In examples, the lower heating control device 14030 may be configured to implement a control methodology based on state of charge of the water heater 2100 - as described herein - to heat the water using AC heating element 14032.

[0234] When not heating, the system charges battery 14010 in anticipation of needing to draw on the stored power under higher loads.12. Interpretation

[0235] The steps of a method, process, or algorithm described in connection with the present disclosure may be embodied directly in hardware, in a software module executed by one or more processors, or in a combination of the two. The various steps or acts in a method or process may be performed in the order shown, or may be performed in another order. Additionally, one or more process or method steps may be omitted or one or more process or method steps may be added to the methods and processes. An additional step, block, or action may be added in the beginning, end, or intervening existing elements of the methods and processes.

[0236] The illustrated embodiments of the disclosure will be best understood by reference to the figures. The foregoing description is intended only by way of example and simply illustrates certain selected exemplary embodiments of the disclosure. It should be noted that the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, apparatuses, methods and computer program products according to various embodiments of the disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which includes at least one executable instruction for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustration, and combinations of blocks in the block diagrams and / or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

[0237] The entire disclosures of all applications, patents and publications cited above and below, if any, are herein incorporated by reference. Reference to any prior art in this specification is not, and should not be taken as, an acknowledgement or any form of suggestion that that prior art forms part of the common general knowledge in the field of endeavour in any country in the world.

[0238] The invention(s) of the present disclosure may also be said broadly to consist in the parts, elements and features referred to or indicated in the specification of the application, individually orcollectively, in any or all combinations of two or more of said parts, elements or features. Where in the foregoing description reference has been made to integers or components having known equivalents thereof, those integers are herein incorporated as if individually set forth.

[0239] Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in at least one embodiment. In the foregoing description, numerous specific details are provided to give a thorough understanding of the exemplary embodiments. One skilled in the relevant art may well recognize, however, that embodiments of the disclosure can be practiced without at least one of the specific details thereof, or can be practiced with other methods, components, materials, et cetera. In other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of the invention.

[0240] Throughout this specification, the word "comprise" or "include", or variations thereof such as "comprises", "includes", "comprising" or "including" will be understood to imply the inclusion of a stated element, integer or step, or group of elements integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps, that is to say, in the sense of "including, but not limited to".

[0241] Aspects of the present disclosure have been described by way of example only and it should be appreciated that modifications and additions may be made thereto without departing from the scope thereof.

Claims

CLAIMS1. A method, including: receiving an indication of flow of liquid from a thermal energy storage device; determining a representation of currently stored energy by the thermal energy storage device, based at least in part on the received indication of flow; and controlling at least one heating source to heat liquid in the thermal energy storage device based at least in part on the determined representation of currently stored energy.

2. The method of claim 1, further including controlling operation of the at least one heating source based at least in part on a demand profile for the thermal energy storage device.

3. The method of claim 1, further including receiving at least one external signal indicative of at least one characteristic of an external operating environment influencing control of the at least one heating source.

4. The method of claim 3, wherein the at least one characteristic is one or more of: amount of curtailment, percentage of renewable energy generation, price, environmental conditions, time of use tariffs, amount of carbon emissions, and network constraints.

5. The method of claim 1, including generating a digital twin of the thermal energy storage device based at least in part on the determined representation of currently stored energy, wherein control of the at least one heating source is based on at least one output from the digital twin.

6. The method of claim 1, wherein the thermal energy storage device is one of a plurality of thermal energy storage devices, and the method further comprises: determining the representation of currently stored energy for each of the plurality of thermal energy storage devices; controlling the respective heating sources of the plurality of thermal energy storage devices based on a cooperative multi-agent control approach.

7. The method of claim 1, wherein the indication of flow is a sensed pressure within a reservoir of the thermal energy storage device.

8. The method of claim 1, including continually determining a current value of the representation of currently stored energy, and adjusting control of the at least one heating source in response.

9. The method of claim 1, wherein the thermal energy storage device is an electric water heater and the at least one heating source is a resistive heating element.

10. A system, comprising:a thermal energy storage device; at least one sensing device configured to output an indication of flow of liquid from the thermal energy storage device; at least one controller configured to: determine a representation of currently stored energy by the thermal energy storage device, based at least in part on the received indication of flow; and control operation of at least one heating source to heat liquid in the thermal energy storage device based at least in part on the determined representation of currently stored energy.

11. A computer-readable storage medium configured with data and with instructions that upon execution by at least one processor will cause the at least one processor to perform a method, comprising: receiving an indication of flow of liquid from a thermal energy storage device; determining a representation of currently stored energy by the thermal energy storage device, based at least in part on the received indication of flow; and controlling operation of at least one heating source to heat liquid in the thermal energy storage device based at least in part on the determined representation of currently stored energy.

Images