Method and system for characterizing thermal energy distribution in a thermal energy exchange system

Abstract

1. A method for characterizing thermal energy distribution in a thermal energy exchange system using a data-driven model, the system having a thermal energy supply unit configured to provide heating / cooling and a plurality of receiving units configured to consume the heating / cooling provided by the thermal energy supply unit. Connections of at least a subset of the plurality of receiving units to a primary supply line and a primary return line of the supply unit are modeled as a single supply line and a single return line, respectively, and values ​​indicative of a temperature and a flow rate of the thermal energy exchange medium in the single supply line and the single return line are unknown to the data-driven model. An input parameter set is provided to a trained machine learning model system configured to output at least one predicted value, the at least one predicted value including values ​​indicative of characteristics of the thermal energy exchange system, the input parameter set including at least a value indicative of a temperature of the thermal energy exchange medium provided by the thermal energy supply unit via the primary supply line.

Description

[Technical Field]

[0001] The present invention relates to a method and a system for characterizing the thermal energy distribution in a thermal energy exchange system using a data-driven model. Furthermore, the present invention relates to a method for controlling the supply temperature of a thermal energy exchange system. Also, the present invention relates to a method for controlling the heat load of one or more further receiving units. Additionally, the present invention relates to a computer program product. [Background technology]

[0002] Modeling of thermal energy distribution in thermal energy exchange systems is essential for various applications or use cases, such as supply temperature control, supply temperature optimization such as supply temperature minimization, heat supply side management, and return temperature control.

[0003] There are various approaches to supply temperature control. These approaches can be based on outdoor temperature compensation, weather compensation, and a heating curve, in which the supply temperature setpoint is determined as a function of the outdoor temperature. This approach assumes that when the outdoor temperature is lower, a higher supply temperature is required, and vice versa. The heating curve is determined from practical experience, and a trade-off is established between supplying sufficient heat to customers and avoiding unnecessarily high supply temperatures accompanied by unnecessarily high network heat losses, while taking into account the flow limitations of the thermal energy distribution network in the thermal energy exchange system.

[0004] A drawback of the heating curve method is that the supply temperature setpoint is determined solely by the outdoor temperature and does not consider the actual current and future needs of the system, which depend not only on the outdoor temperature but also on the consumer's behavior and the characteristics of the consumer's local thermal energy system. Furthermore, the heating curve method focuses only on meeting heat demand and minimizing heat loss. Therefore, other operational goals, such as dynamic optimization of the heat supply profile, are not enabled by this method. To overcome these drawbacks, operators may manually temporarily increase or decrease the supply temperature setpoint using their practical experience.

[0005] These manual interventions can be automated by computer-based approaches, such as supply temperature optimization, in which the supply temperature setpoint is typically determined using a numerical optimization solver to dynamically calculate its value based on current operating conditions, predictive mathematical models, and custom target measurement data.

[0006] White-box models are often used to characterize / model the thermal energy distribution in thermal energy exchange systems. Such white-box models can be derived from physical laws using the physical properties of the system. Analytical models (see so-called white-box models), which are also used for digital twins, are often not suitable because systems can be very complex, resulting in many parameters that are difficult to accurately determine.

[0007] Often, some technical input data required for analytical or white-box models related to the network layout, thermal network configuration, component dimensions and locations, insulation properties, etc. of the thermal energy exchange system may be unavailable and / or unknown. Additionally, sensors deployed to measure specific properties at various locations in the thermal network may not exist. Collecting that data may be very labor-intensive and / or expensive, and some information may be difficult to obtain or difficult to accurately determine, which may lead to inaccuracies in the system modeling. It may be difficult to collect relevant data, especially in real time, to be able to perform appropriate control of the thermal energy exchange system.

[0008] For example, for district heating systems, a relatively large amount of data / information needs to be available to the model. Such data can be difficult to collect and unstructured. Furthermore, model setup and maintenance can require specific expertise and be time-consuming. Furthermore, because they are based on physical principles and measurable properties, models tend to be computationally complex, i.e., have a large number of equations and variables. This is not convenient for optimal control purposes, for example, where the model should be quickly simulated. There is a strong desire to improve the characterization of thermal energy distribution in thermal energy exchange systems, especially for relatively complex systems.

[0009] Due to the complexity of using these analytical models, it can be advantageous to use empirical models (see black-box models), which can utilize machine learning models (e.g., regression models) to make predictions. Such models may, for example, model the supply temperature propagation from the heat production site to the monitored building, the return temperature propagation from all buildings to the heat production site, the collective flow rate of unmonitored buildings, etc. Data-driven trained models can be derived from measured data using mathematical regression techniques. Hybrid models (gray-box) are also used in some cases, where the model equations are inspired by physical laws, but the model parameters are determined to fit the measured data. Summary of the Invention

[0010] It is an object of the present invention to provide a method and system that eliminates at least one of the above-mentioned drawbacks.

[0011] Additionally or alternatively, it is an object of the present invention to improve the characterization of thermal energy distribution in thermal energy exchange systems.

[0012] Additionally or alternatively, it is an object of the present invention to more efficiently model thermal energy exchange systems for purposes of control of the thermal energy exchange system, such as supply temperature control, heat supply side management, etc.

[0013] Therefore, the present invention provides a computer-implemented method for characterizing thermal energy distribution in a thermal energy exchange system using a data-driven model, the thermal energy exchange system comprising a thermal energy supply unit configured to provide heating / cooling and a plurality of receiving units configured to consume the heating / cooling provided by the thermal energy supply unit, the thermal energy supply unit being connected to a primary supply line through which a thermal energy exchange medium flows from the thermal energy supply unit towards the plurality of receiving units, and the thermal energy supply unit being connected to a primary return line through which the thermal energy exchange medium from the plurality of receiving units flows towards the thermal energy supply unit. a data-driven model of a thermal energy exchange system, the data-driven model comprising: a thermal energy exchange system configured to generate a thermal energy exchange medium flowing toward the thermal energy exchange system; a thermal energy exchange medium flowing toward the thermal energy exchange system; and connections of at least a subset of the plurality of receiving units to a primary supply line and a primary return line modeled as a single supply line and a single return line, respectively; values ​​indicative of a temperature and a flow rate of the thermal energy exchange medium in the single supply line and the single return line being unknown to the data-driven model; the method comprising providing an input parameter set to a trained machine learning model system configured to output at least one predicted value, the at least one predicted value comprising a value indicative of a characteristic of the thermal energy exchange system, the input parameter set comprising at least a value indicative of a temperature of the thermal energy exchange medium supplied by the thermal energy supply unit via the primary supply line.

[0014] Advantageously, the method can effectively handle cases where various receiving units have connections in the thermal network with unknown / unmonitored parameters (e.g., temperature and / or flow rate). The thermal network of a thermal energy exchange system may include various pipes, units, and subsystems arranged to enable thermal energy exchange. The method can also be used when the receiving units do not have known / monitored parameters, for example, all are unmonitored receiving units.

[0015] The thermal energy distribution in a thermal energy exchange system depends at least in part on data (e.g., the temperature of the thermal energy exchange medium) that is unknown / unavailable to the machine learning model at connections or points within the thermal energy exchange system. For example, the unknown / unavailable data may be the return temperature at an unmonitored connection of a receiving unit. The term "unmonitored" may indicate not being monitored by the model or that the relevant data is unavailable and / or unknown. For example, for supply temperature control, a data-driven trained machine learning model may be used to model the effect of the supply temperature of a heating / cooling supply side on the return temperature of that heating / cooling supply side. The thermal energy exchange system may be modeled such that a subset of receiving units are grouped together, and their connections to the primary supply line and primary return line may be bundled as a single supply line and a single return line, respectively.

[0016] A single supply line may model the flow of thermal energy exchange medium from the thermal energy supply unit toward a subset of the plurality of receiving units, and a single return line may model the flow of thermal energy exchange medium from the subset of the plurality of receiving units toward the thermal energy supply unit, in some examples, the subset of receiving units does not have a metered connection available to the model.

[0017] The method can be applied with very few requirements when it is applied, and it only requires data on a small number of parameters. Data on other parameters and other information is not required. Even if such data or information is available, it does not need to be managed (requested, collected, processed, stored), which can save time and effort in characterizing a thermal energy exchange system.

[0018] The present invention uses data-driven models to enable improved operational control of thermal energy exchange systems.

[0019] Different control objectives may be used, such as supply temperature minimization or heat supply optimization. For example, heat supply optimization may relate to improving any thermal-hydraulic characteristic associated with the heat supply unit. The characteristic may be, for example, heat load, return temperature, flow rate, heat production cost, greenhouse gas emissions, etc. In some examples, other thermal-hydraulic characteristics may be calculated using return temperature, supply temperature, and / or flow rate.

[0020] The method allows for predicting the effect of the control variables on the control targets, which provides information for determining the set points and / or control signals to be applied to the actual thermal energy exchange system.

[0021] Data-driven or machine learning models are built using historical data. Models can be trained using training data. Model parameter configuration can be automated without the need for manual configuration, which can typically be labor-intensive. Model configuration relies on historical measurement data instead of detailed information about system components. The former can easily be collected from the same sensors needed for control.

[0022] A district heating system that makes up a heat network may include one or more supply units (e.g., heat plants) and receiving units (e.g., consumer units such as buildings). However, intermediate exchange units (e.g., “substations” housing heat exchangers) may also exist. Such intermediate exchange units may, for example, have a heat exchanger having a primary side and a secondary side. Physical separation between fluids flowing on either side may be provided. For example, heat / cooling on the primary side may be exchanged on the secondary side of the intermediate exchange unit. It will be understood that in some embodiments, such intermediate units may be considered receiving units. The receiving unit may be a consumer unit and / or an intermediate unit between the supply side and another intermediate unit or consumer units. However, in some embodiments, such intermediate units are not modeled. For example, the supply side may be indirectly connected to an end receiving unit via one or more intermediate units, such as heat exchangers.

[0023] A thermal energy receiving unit may be understood as an element in a thermal system that receives heat, either for direct use or for further distribution, for example a building / apartment connection and / or a network substation (intermediate exchange unit).

[0024] Heat supply units may be understood as elements that deliver heat to a thermal energy exchange system. These may be heat production units, but are not limited to, boilers (gas, oil, biomass, fuel, etc.), heat pumps, combined heat and power (CHP) units, etc. Furthermore, in some embodiments, they may be heat transfer units that transfer heat coming from other thermal energy exchange systems, either hydraulically connected or not, with or without the use of heat exchangers.

[0025] For supply temperature propagation, a supply temperature machine learning model (e.g., a regressor) can be used to predict the outlet temperature of the primary supply line using the inlet temperature of the primary supply line. The data-driven model can be used not only to calculate the supply temperature propagation, but also to predict the best supply temperature for the system according to the target.

[0026] For example, for supply temperature control / optimization, the physical behavior of the network plays an important role (partly due to the time delays that occur). According to the method of the present invention, it is not necessary to have a priori technical information about the layout, location, diameter, length, insulation properties, arrangement of the thermal network, etc.

[0027] In some embodiments, supply temperature control may be performed based on, for example, an energy peak shaving control target. The method of the present invention may be used to characterize / model supply temperature propagation from the supply side (referring to the thermal energy supply unit) to the demand side (referring to the receiving unit), how return temperature propagates to the supply side, the effect of supply temperature on return temperature and / or flow rate coming from the receiving unit, etc.

[0028] For supply temperature control, where the control objective is supply temperature minimization (which may be referred to as supply temperature minimization), it may be necessary to predict the supply temperature at the monitored location to allow for lowering the inlet supply temperature while taking into account limitations on the outlet supply temperature. For supply side management, the control objective is optimization of a dynamic heat supply profile, and depending on the selection of the supply temperature, it may be necessary to predict the return temperature and flow rate at the heat supply unit, which are affected by both known / monitored and unknown / unmonitored connections. In certain cases, only one of the temperature or flow rate, or another quantity (e.g., heat load), may be sufficient.

[0029] In some embodiments, heat load management can be performed as long as control constraints on supply temperature are defined, for example, based on outdoor temperature. However, for supply temperature minimization, at least one "control point," and therefore a monitored receiving unit, may be required.

[0030] Optionally, the value indicative of a characteristic of the thermal energy exchange system is at least one of a value indicative of a return temperature of the thermal energy exchange medium provided to the thermal energy supply unit via the primary return line, or a value indicative of a flow rate of the thermal energy exchange medium in the single supply line, the single return line, the primary supply line and / or the primary return line.

[0031] The return temperature is a function of the individual temperatures of the thermal energy exchange media leaving the multiple receiving units toward the thermal energy supply unit. The return temperature is often a key variable that affects the efficiency, cost, and / or sustainability of the heat supply. For example, heat supply units that rely on combustion processes have higher fuel efficiency when lower return temperatures allow flue gas condensation. For example, geothermal sources allow higher production capacity and shorter capital investment payback times when return temperatures are low.

[0032] To improve the characterization of the system, a mass balance of the entire network can be calculated. For mass balance, it may be necessary to find the flow from one point in the thermal network towards other branches.

[0033] Generally, the flow rate in a thermal network has a strong influence on the travel time of the heating medium flowing through the network. This causes time delays that strongly affect the propagation of temperature through the network. Generally, the ability to predict the flow rate in the system, specifically the flow rate that can be measured or identified at the connection to the network, allows for better prediction of other relevant variables.

[0034] Furthermore, the instantaneous heat output of the heat supply unit depends on the return temperature and the flow rate, which makes it particularly appropriate to be able to predict these values.

[0035] Optionally, the set of input parameters includes a value related to the thermal load of the receiving unit.

[0036] In some examples, the thermal loads of the receiving units may be unknown for at least a subset of the receiving units and therefore are not used as direct inputs. Instead, known values ​​useful for predicting the thermal loads may be used without taking the intermediate step of predicting the thermal loads to be used as inputs for another model. However, in some examples, the thermal loads are predicted by a thermal load prediction model (e.g., a further trained machine learning model).

[0037] The heat load of an individual receiving unit, or an aggregated set of receiving units, is a variable that has a strong and determinative influence on the flow rates and return temperatures in the thermal network. If representative values ​​can be provided as input to the method, this will improve the quality of the prediction.

[0038] Optionally, the value related to the thermal load of the receiving unit is calculated based on a value indicating a time, such as the time of day.

[0039] Optionally, the value relating to the heat load of the receiving unit is calculated based on a value indicative of weather, such as outdoor temperature.

[0040] The supply return temperature may depend on the heat demand of the receiving unit and the supply temperature that may be measured at the receiving unit. Heat demand may be linked to weather (e.g., weather forecast, outdoor temperature, humidity, etc.) and time (e.g., time of day, day of week).

[0041] Heat demand can be correlated with weather. For example, outside temperature can be one of the most important parameters affecting heat demand. Hot water usage (e.g., showers, heating systems being turned off at night, etc.) is often linked to the time of day. Thus, heat demand not only depends on the outside temperature, but also varies with the time of day. The day of the week can also have an impact (e.g., weekends, holidays, etc.).

[0042] Therefore, using weather and / or time related values ​​as inputs to the method allows for more accurate predictions that better capture the inherent variability in the operation of heat distribution systems.

[0043] Optionally, the system is provided with one or more further receiving units, each having a respective supply line and a respective return line connected to the primary supply line and the primary return line, and the temperature and / or flow rate of the thermal energy exchange medium in the respective supply line and the respective return line is known to the model.

[0044] Optionally, the at least one predicted value further comprises a value indicative of a supply temperature of the thermal energy exchange medium in an individual supply line of the further receiving unit.

[0045] Optionally, the set of input parameters includes at least one of a value indicating the flow rate of the thermal energy exchange medium in the individual supply lines and / or individual return lines of the further receiving unit, a value indicating the flow rate of the thermal energy exchange medium in the primary supply line and / or primary return line, or a value indicating the return temperature in the individual supply line and / or individual return line of the further receiving unit.

[0046] Optionally, the flow rate of the thermal energy exchange medium in the primary supply line and / or the primary return line is calculated based on values ​​indicative of the flow rate of the thermal energy exchange medium in the single supply line and / or the single return line.

[0047] Optionally, one or more previous predictions are provided in the input parameter set.

[0048] Optionally, the value indicative of the temperature of the thermal energy exchange medium supplied by the thermal energy supply unit via the primary supply line is measured indirectly using the measured temperature of the thermal energy exchange medium in the respective supply line of the further receiving unit.

[0049] Optionally, the input parameter set comprises values ​​indicative of the heat load of the further receiving unit, the flow rate in the further receiving unit, and / or the return temperature of the further receiving unit.

[0050] According to an aspect, the present invention provides a system for characterizing thermal energy distribution in a thermal energy exchange system using a data-driven model, the thermal energy exchange system comprising: a thermal energy supply unit configured to provide heating / cooling; and a plurality of receiving units configured to consume the heating / cooling provided by the thermal energy supply unit, the thermal energy supply unit connected to a primary supply line through which a thermal energy exchange medium flows from the thermal energy supply unit towards the plurality of receiving units; the thermal energy supply unit connected to a primary return line through which the thermal energy exchange medium flows from the plurality of receiving units towards the thermal energy supply unit; The system provides a system in which the connections of at least a subset of a plurality of receiving units to a primary supply line and a primary return line are modeled as a single supply line and a single return line, respectively, and values ​​indicative of the temperature and flow rate of the thermal energy exchange medium in the single supply line and the single return line are unknown to the data-driven model, the system including a processor configured to provide a set of input parameters to a trained machine learning model system configured to output at least one predicted value, the at least one predicted value including a value indicative of a characteristic of the thermal energy exchange system, the input parameter set including at least a value indicative of the temperature of the thermal energy exchange medium supplied by the thermal energy supply unit via the primary supply line.

[0051] The system can be used to perform supply temperature control of a thermal network, such as supply temperature minimization and supply-side management (heat load, temperature, price, and operating constraints). By using data-driven modeling, model parameter configuration can be automated. There may be no need for manual configuration, which requires a time investment by an expert. Model configuration relies on historical measurement data instead of detailed information about system components. While the former can be easily collected from the same sensors needed for control, the latter requires communication between the solution provider and the user. Furthermore, such information can be scattered and difficult to collect. The method can also accurately calculate / predict flow rates of unmonitored receiving units (e.g., buildings or other consumer units).

[0052] Optionally, the system is used for the control of one or more reception units, and thus the system can be used for demand side management.

[0053] The machine learning model may be trained by collecting historical sensor data to be used as training data. One or more test campaigns may be run to train the data-driven model.

[0054] According to an aspect, the present invention provides a computer-implemented method for controlling the thermal load of one or more further receiving units, wherein at least one predicted value associated with the thermal energy exchange system has been predicted using the method according to the present invention, and the control of the thermal load of the one or more further receiving units is performed based on a predetermined target goal and the at least one predicted value.

[0055] Supply temperature control is one element of the operational control system in a district heating heat network. The goal of this subsystem is to determine an improved setpoint (see optimization) for the supply temperature provided by the heat production units, i.e., the temperature of the heat exchange medium in the primary supply line. The optimal supply temperature setpoint depends on the real-time conditions and requirements (e.g., customer / consumer requirements) from the receiving units. In addition to the primary goal of ensuring that heat demand can be met, this would allow achieving secondary goals, such as supply temperature minimization, i.e., reducing the temperature in the supply pipes also reduces the heat loss in these pipes to the environment, and / or activating the flexibility of the district heating network, since dynamic control of the supply temperature allows a partial decoupling between the heat demand (determined by the customer) profile and the heat supply (provided by the heat production units) profile. This allows for shaping the heat supply profile in order to optimize the heat production process (e.g., cost / revenue optimization, GHG emission reduction).

[0056] In some embodiments, a supervisory control system is used for supply temperature control, possibly integrated with the control of other operating variables, which can determine the optimum supply temperature (with respect to a given control target) dynamically, i.e., as a function of time, and with continuous feedback of the actual conditions in the district heating system. The method according to the invention is used to accurately and effectively model / predict flow rates and temperatures in a thermal network.

[0057] According to an aspect, the present invention provides a computer-implemented method for controlling the thermal load of one or more further receiving units, wherein at least one predicted value associated with the thermal energy exchange system has been predicted using the method according to the present invention, and the control of the thermal load of the one or more further receiving units is performed based on a predetermined target goal and the at least one predicted value.

[0058] It will be understood that a receiving unit, such as a building, may refer to a connection in the thermal network of a system. It does not strictly have to be a building, or even a consumer unit. It can be any (sub)system element that draws flow from a supply line and returns it to a return line. A receiving unit can also be, for example, a substation that serves one or more receiving units, such as a building, a connection to a downstream thermal network, etc. A "connection" can be understood as an interface between lines (e.g., pipes) corresponding to the primary and secondary networks, respectively. This interface can be characterized by ownership, responsibility, network topology / hierarchy, measurement location, with or without a physical line / pipe joint.

[0059] It will be understood that a "monitored connection" may refer to a connection for which at least a supply temperature measurement is available to the model. Connections that do not meet the definition of a "monitored connection" may be considered "unmonitored connections." In accordance with the present invention, unmonitored connections are not individually identified, but are modeled only in a collective manner.

[0060] In some embodiments, the receiving units are not modeled directly, but rather the distribution of heat through a thermal network that includes the receiving units. The thermal network can have a variety of configurations.

[0061] A machine learning model may have various inputs / outputs. Input data may be provided to the model, and output data may be provided as outputs by the model. It will be understood that the inputs and outputs mentioned may only define the variables used, without specifying whether this relates to past, present, or future values.

[0062] It will be understood that any of the aspects, features, and options described for the method apply equally to the system, computer program product described, and it will also be apparent that any one or more of the above aspects, features, and options may be combined. [Brief explanation of the drawings]

[0063] The invention will be further explained on the basis of exemplary embodiments represented in the drawings, which are shown by way of non-limiting illustration, and it should be noted that these figures are merely schematic representations of embodiments of the invention, shown by way of non-limiting example.

[0064] [Figure 1] 1 shows a schematic diagram of an exemplary embodiment of a thermal energy exchange system; [Figure 2] 1 shows a schematic diagram of an exemplary method. DETAILED DESCRIPTION OF THE INVENTION

[0065] 1 shows a schematic diagram of an exemplary embodiment of a thermal energy exchange system 1. In this example, the system 1 is a district heating (see collective heating) system that constitutes a thermal network. Such networks can be very complex, and only a simplified representation is shown to illustrate the method according to the invention.

[0066] According to the present disclosure, thermal energy distribution in a thermal energy exchange system is characterized using a machine learning model. The thermal energy exchange system 1 includes a thermal energy supply unit 3 configured to provide heating / cooling. The system 1 further includes a plurality of receiving units configured to consume the heating / cooling provided by the thermal energy supply unit 3. In this example, the system 1 includes a plurality of unmonitored connections 5b and a plurality of optional monitored connections 5a. However, in some examples, only the unmonitored connections 5b are available.

[0067] The thermal energy supply unit 3 is connected to a primary supply line 7 through which the thermal energy exchange medium flows from the thermal energy supply unit 3 towards the plurality of receiving units 5. The thermal energy supply unit 3 is connected to a primary return line 9 through which the thermal energy exchange medium flows from the plurality of receiving units 5 towards the thermal energy supply unit 3. The connections 5b of at least a subset of the plurality of receiving units 5 to the primary supply line 7 and primary return line 9 are modeled as a single supply line 11 and a single return line 13, respectively, and values ​​indicative of a temperature and a flow rate of the thermal energy exchange medium in the single supply line 11 and the single return line 13 are unknown to the data-driven model. The method includes providing an input parameter set to a trained machine learning model system configured to output at least one predicted value, the at least one predicted value including a value indicative of a characteristic of the thermal energy exchange system 1, the input parameter set including at least a value indicative of a temperature of the thermal energy exchange medium supplied by the thermal energy supply unit 3 via the primary supply line 7.

[0068] In some embodiments, the value indicative of a characteristic of the thermal energy exchange system is at least one of a value indicative of a return temperature of the thermal energy exchange medium provided to the thermal energy supply unit via the primary return line, or a value indicative of a flow rate of the thermal energy exchange medium in the single supply line, the single return line, the primary supply line and / or the primary return line.

[0069] The method according to the present invention can be used in combination with various use cases. Some examples are provided below. It is understood that in various examples, it is also envisioned that a cooling supply unit is provided instead of a heat supply unit. Thus, in some alternative examples, whenever "heat" is mentioned, the heat can be "cooling." It is also envisioned that the system can be switched between a heating state and a cooling state.

[0070] The modeling / characterization method may be used in combination with supply temperature control, where the variable for supply temperature control may be the supply temperature of the heat supply unit.

[0071] In some embodiments, supply temperature minimization is performed. The control objective can be considered as minimizing the supply temperature of the heat supply unit while respecting the minimum supply temperature at a specific downstream location. The system can have at least one heat supply unit, at least one "monitored" heat receiving unit, and a thermal network with at least supply and return lines / pipes connecting the at least one "monitored" heat receiving unit with the at least one heat supply unit. The output of the model can be the (supply) temperature of at least one downstream line location, indicating the (supply) temperature requirement of the heat receiving unit. The input to the model can be the supply temperature of the heat supply unit. It will be understood that a "monitored heat receiving unit" can be understood as any heat receiving unit for which an indicative supply temperature is available / known / used to / by the model.

[0072] For feed temperature minimization, the feed temperature may be selected as low as possible so that the feed temperature at the receiving unit is maintained above a minimum threshold.

[0073] The system may have at least a number of "unmonitored" (ie, unknown / unavailable to the model) heat receiving units.

[0074] Optionally, the system has additional "monitored" heat receiving units.

[0075] Additionally or alternatively, the system includes an additional heat supply unit.

[0076] In some embodiments, the output indicates the supply temperature at the additional "monitored" heat receiving unit. Additionally or alternatively, the output indicates the flow rate at the heat supply unit and / or the additional heat supply unit. The flow rate at the heat supply unit (assuming it is the only heat supply unit) is the sum of the flow rates of all monitored heat receiving units, assumed to be predicted externally, which can advantageously be predicted using a machine learning model, and the collective flow rates of all unmonitored heat receiving units. Additionally or alternatively, the output indicates the total flow rate (sum of the flow rates) of the unmonitored heat receiving units.

[0077] In some embodiments, the input indicates a flow rate at the heat supply unit and / or the additional heat supply unit. Additionally or alternatively, the input indicates a flow rate at the location receiving unit and / or the additional "monitored" heat receiving unit. Additionally or alternatively, the input indicates a total flow rate (sum of flow rates) for unmonitored heat receiving units. Additionally or alternatively, the input indicates a supply temperature at the receiving unit and / or the additional heat receiving unit. Additionally or alternatively, the input indicates a supply temperature of the additional heat supply unit. Additionally or alternatively, the input indicates an ambient / environmental variable such as ground surface temperature. Additionally or alternatively, the input indicates a weather variable such as outside temperature. Additionally or alternatively, the input indicates a time, e.g., time of day, day of the week, day of the year.

[0078] According to the present invention, a distinction is made between heat receiving units whether or not the supply temperature is monitored. Flow rate in the present system can be an optional input for modeling. The accuracy of the modeling / characterization can be significantly improved by also taking flow rate into account. For a subset of heat receiving units, flow rate can be predicted collectively rather than individually.

[0079] Additionally or alternatively, the modeling / characterization method may be used for heat supply optimization, where the control objective may be improvement of any thermal-hydraulic characteristic associated with the heat supply unit, such as heat load, return temperature, flow rate, etc. The method may be used for any thermal-hydraulic characteristic that requires data indicative of return temperature and / or flow rate to be calculated.

[0080] The system may have at least one heat supply unit, at least a number of unmonitored heat receiving units, and optionally one or more monitored heat receiving units.

[0081] The output may indicate thermal hydraulic characteristics of the heat supply unit, such as at least one of the return temperature in the primary return line, and / or the flow rate in the primary supply line, and / or any other thermal hydraulic characteristics that depend on the return temperature and / or flow rate, such as, for example, the thermal load (Q=m*cp*(Tsup-Tret)), which is the rate of energy delivered by the heat supply unit per unit time, and / or the primary energy consumption of the heat supply unit (e.g., PE=Q / efficiency, where efficiency may depend on Tsup and / or Tret).

[0082] The input may indicate the supply temperature at the heat supply unit in the primary supply line.

[0083] Optionally, the system may include additional heat receiving units and / or additional heat supply units. In some embodiments, the output indicates the supply temperature at one or more "monitored" heat receiving units. Additionally or alternatively, the output indicates the flow rate at the heat supply unit (in the primary return line or in the primary supply line) and / or at the additional heat supply units. Additionally or alternatively, the output may indicate the total flow rate, i.e., the sum of the flow rates, of the heat receiving units not being monitored.

[0084] In some examples, the input indicates a supply temperature at one or more "monitored" heat receiving units. Additionally or alternatively, the input indicates a return temperature at one or more "monitored" heat receiving units. Additionally or alternatively, the input indicates a flow rate at the heat supply unit (in the primary supply line or primary return line) and / or at an additional heat supply unit. Additionally or alternatively, the input indicates a flow rate at one or more "monitored" heat receiving units. Additionally or alternatively, the input indicates a total flow rate (sum of flow rates) at unmonitored heat receiving units. Additionally or alternatively, the input indicates a supply temperature at an additional heat supply unit. Additionally or alternatively, the input indicates an ambient variable, such as ground temperature. Additionally or alternatively, the input indicates a weather variable, such as outside temperature. Additionally or alternatively, the input indicates a time, e.g., time of day, day of the week, day of the year. Additionally or alternatively, the input indicates thermal-hydraulic characteristics at the heat supply unit.

[0085] For certain goals, it may be necessary to be able to predict the flow rates in the heat supply units, which may be preferable to improve the accuracy of the model predictions. This can be predicted directly or as the sum of the flow rates of all heat receiving units. In the latter case, some individual flow rates can be predicted using methods known in the art, while others may need to be predicted collectively as part of the method according to the invention.

[0086] The modeling method can be used for demand side management, where the variable for demand side management can be the thermal load of the heat receiving units of the system.

[0087] The system may include at least one heat supply unit and at least one "controllable" heat receiving unit. The outputs may be the same as those for heat supply optimization. For example, the outputs may indicate a thermal hydraulic characteristic of the heat supply unit, such as the return temperature in the primary return line and / or the flow rate in the primary return line or primary supply line, and / or any other thermal hydraulic characteristic that depends on the return temperature and / or flow rate (e.g., the heat load, which may be defined as the rate of energy delivered by the heat supply unit per unit of time).

[0088] The inputs may indicate at least one controllable heat load (heat output). Additionally or alternatively, the inputs may indicate a primary flow rate, a primary return temperature, a secondary supply temperature, an outdoor temperature offset, etc.

[0089] Modeling example A district heating network configured to transport a heat exchange medium between a heat producing site and a heat consumer. In this example, the following assumptions are made: a single heat producing site, one or more heat consumers with measured supply and return temperatures and flow rates, and one or more heat consumers without complete measurements of available supply and return temperatures and flow rates (to the model).

[0090] In this example, the district heating network model consists of the following submodels: 1. Predicting the aggregate flow rate of unmonitored connections

number

number

[0091] The second sub-model represents the conservation of mass in the district heating network.

number

[0092] If there is only one connection to monitor,

number

[0093] Other relationships are empirically modeled using data from historical measurements. This results in explicit formulas for the associated output variables, also called "predictors," "regression models," "regressors," "black-box models," or "machine learning models." The selection of input variables is considered first. The functional forms relating inputs to the outputs of these regressors are considered later.

[0094] Note that the following conversion relationship between mass flow rate and volumetric flow rate holds, where ρ is the mass density of the water in the pipe:

number

number

number

[0095] Input variables The regression model calculates the number of past time steps from the previous time step. p Several input variables may be used, including data up to

[0096] Regressors for the aggregate flow rate of unmonitored connections may include:

number

[0097] Output variable F p,0 The previous value of is used as an input to take into account inertial effects. In some embodiments, only the previous value is used, and not the old value.

[0098] The flow rate of each individual unmonitored connection will depend on the primary supply temperature and the substation heat load. The primary supply temperature is not measured directly, but is the heating supply temperature T ss,hs The primary supply temperature may also depend on the flow rate of an exemplary monitored building. In some examples, the regressor for Fp,0 may be directly replaced by a regressor for Fs,hs (directly incorporating mass conservation). This is made possible by the availability of the monitored flow rate as an input feature. Heat loads are not measured but are time-dependent and are thought to be empirically correlated with weather. These are measured over the course of a day. d , minutes of the hour m h , and outdoor temperature T o,fc It is represented in a simplified manner by

[0099] Therefore, the total flow rate of all unmonitored connections can be assumed to depend on the same input variables as the individual connections.

[0100] Regressor for supply temperature on connection 1:

number

[0101] The following physical phenomena are assumed to occur: Convective time delay: ■ Time delay determined by the velocity (flow rate) in the pipe section passing through the track between the heat supply unit and the connection 1. F p,1 and F s,hsIt is assumed that knowledge of is sufficient, i.e., the internal distribution of flow rates within the set of unmonitored connections does not play a significant role. Under this assumption, the flow rates of all (intermediate) pipes in the network can directly depend on the available data. ■Past inlet temperature T ss,hs Thermal inertia of the heating medium and pipe walls: ■Outlet supply temperature T ps,1,k acts as a state variable that "remembers" the previous state Pipe wall heat loss to the environment: ■ The data or model does not include an explicit land surface temperature variable. However, if heat loss does occur, this is reflected in the model parameters, so the outlet temperature will be lower than the inlet temperature.

[0102] Regressor for the return temperature of the heat supply unit:

number

[0103] The same physical phenomena occur in the propagation of the return temperature from the connection 1 to the heat supply unit network as in the supply network.

[0104] In addition, there is also a mix of return temperatures from unmonitored connections. However, these return temperatures are not measured directly. They are assumed to depend on the same variables as the lump-sum flow rate of the unmonitored connections.

[0105] The number of past time steps used, N p can be based on data analysis including (partial) autocorrelation functions, focusing on the supply temperature propagation problem. The outlet supply temperature was found to depend on at most n time steps prior to the inlet supply temperature. It will be understood that n is case-dependent (e.g., 5 or 6 time steps out of 10 minutes, however, various other values ​​can be used). The same or different numbers can be used for all other input variables and regression models.

[0106] Approaches for extending to multiple monitored buildings: In some embodiments, for a regressor for the lumped flow of unmonitored buildings, iの流量F p,i,can be added as an additional input (and optionally a past value).

[0107] Regressors for supply temperature: For each monitored connection i, the input can be: ■T ps,i :Recent N p value ■T ss,hs :Recent N p value ■F p,j : The most recent N for each monitored connection j p Values ​​(including i itself) ■F p,hs :Recent N p value Regressors for return temperature: For each monitored connection i, the following inputs may be included: ■T pr,i :Recent N p value ■F p,i :Recent N p value

[0108] Note that in principle it is also possible to model the entire network without monitored connections using variations of the machine learning models (e.g. regressors) for unmonitored flow rates and return temperatures.

[0109] FIG. 2 shows a schematic diagram of an exemplary method for characterizing thermal energy distribution in a thermal energy exchange system using a machine learning model. The thermal energy exchange system includes a thermal network with a specific layout / topology that is unknown. A thermal energy supply unit provides heating / cooling, and multiple receiving units consume the heating / cooling provided by the thermal energy supply unit. The thermal energy supply unit is connected to a primary supply line through which a thermal energy exchange medium flows from the thermal energy supply unit to the multiple receiving units. The thermal energy supply unit is connected to a primary return line through which the thermal energy exchange medium flows from the multiple receiving units to the thermal energy supply unit. The connections of at least a subset of the multiple receiving units to the primary supply line and primary return line are modeled as a single supply line 11 and a single return line 13, respectively, and values ​​indicative of the temperature and flow rate of the thermal energy exchange medium in the single supply line and single return line are unknown to the data-driven model. Three separate lines are shown between the thermal network and the "unmonitored" heat receiving unit, for both supply and return, respectively, but these lines are bundled together during modeling. The method includes providing a set of input parameters to a trained machine learning model system configured to output at least one predicted value, the at least one predicted value including a value indicative of a characteristic of the thermal energy exchange system, the input parameter set including at least a value indicative of a temperature of the thermal energy exchange medium supplied by the thermal energy supply unit via the primary supply line.

[0110] In this exemplary embodiment, multiple optional "monitored" heat receiving units are also shown, with connections being modeled separately without bundling for "unmonitored" heat receiving units.

[0111] For supply temperature control, the value of the supply temperature can be controlled to reach a specific goal. For example, supply temperature minimization can be performed, in which the supply temperature is controlled with the goal of making the supply temperature as low as possible, subject to the constraints of the receiving unit (e.g., a consumer unit such as a building). Other supply temperature optimizations can also be performed, for example, based on the thermal load of the heat supply unit.

[0112] It will be understood that the method may include computer-implemented steps. All steps described above may be computer-implemented steps. Embodiments may comprise a computer apparatus, and the process may be executed on a computer apparatus. The invention also extends to computer programs adapted for carrying out the invention, in particular computer programs on or in a carrier. The program may be in the form of source code or object code, or any other form suitable for use in carrying out the process according to the invention. The carrier may be any entity or device capable of carrying a program. For example, the carrier may comprise a storage medium such as a ROM, e.g., a semiconductor ROM or a hard disk. Furthermore, the carrier may be a transmissible carrier such as an electric or optical signal, which may be conveyed via an electric or optical cable, or by radio or other means, e.g., via the Internet or the cloud.

[0113] Some embodiments may be implemented, for example, using a machine or tangible computer-readable medium or article that may store instructions or sets of instructions that, when executed by the machine, cause the machine to perform methods and / or operations in accordance with the embodiments.

[0114] Various embodiments may be implemented using hardware elements, software elements, or a combination of both. Examples of hardware elements may include a processor, a microprocessor, a circuit, an application specific integrated circuit (ASIC), a programmable logic device (PLD), a digital signal processor (DSP), a field programmable gate array (FPGA), a logic gate, a register, a semiconductor device, a microchip, a chipset, etc. Examples of software may include a software component, a program, an application, a computer program, an application program, a system program, a machine program, an operating system software, a mobile app, middleware, firmware, a software module, a routine, a subroutine, a function, a computer-implemented method, a procedure, a software interface, an application program interface (API), a method, an instruction set, computing code, computer code, etc.

[0115] The present invention has been described herein with reference to specific examples of embodiments of the invention. However, it will be apparent that various modifications, variations, substitutions, and alterations can be made therein without departing from the essence of the invention. For purposes of clarity and concise description, features are described herein as part of the same or separate embodiments; however, alternative embodiments having all or some combinations of the features described in these separate embodiments are also contemplated and understood to fall within the framework of the invention as outlined by the claims. Accordingly, the specification, drawings, and examples should be regarded in an illustrative, rather than a restrictive, sense. The present invention is intended to embrace all such alternatives, modifications, and variations that are within the scope of the appended claims. Furthermore, many of the described elements are functional entities that can be implemented as separate or distributed components, or in combination with other components, in any suitable combination and location.

[0116] In the claims, any reference signs placed between parentheses shall not be interpreted as limiting the claim. The word "comprising" does not exclude the presence of other features or steps than those listed in a claim. Furthermore, the words "a" and "an" shall not be interpreted as limiting to "one and only one" but are instead used to mean "at least one" and do not exclude a plurality. The term "and / or" includes any and all combinations of one or more of the associated listed items. The mere fact that certain measures are recited in mutually different claims does not indicate that a combination of these measures can be used to advantage. [Table 1]

Claims

1. A computer-aided method for characterizing the thermal energy distribution in a thermal energy exchange system using a data-driven model, wherein the thermal energy exchange system comprises a thermal energy supply unit configured to supply heating / cooling, and a plurality of receiving units configured to consume the heating / cooling supplied by the thermal energy supply unit, the thermal energy supply unit being connected to a primary supply line through which a thermal energy exchange medium flows from the thermal energy supply unit to the plurality of receiving units, the thermal energy supply unit being connected to a primary return line through which a thermal energy exchange medium flows from the plurality of receiving units to the thermal energy supply unit, and at least a subset of the plurality of receiving units The method comprises providing an input parameter set to a trained machine learning model system configured to output at least one predicted value, the at least one predicted value comprising a value representing the characteristics of the heat energy exchange system, and the input parameter set comprising at least a value representing the temperature of the heat energy exchange medium supplied by the heat energy supply unit via the primary supply line.

2. The method according to claim 1, wherein the value indicating the characteristics of the heat energy exchange system is at least one of the following: a value indicating the return temperature of the heat energy exchange medium provided to the heat energy supply unit via the primary return line; a value indicating the flow rate of the heat energy exchange medium in the single supply line, the single return line, the primary supply line, and / or the primary return line; or a value calculated based on the return temperature and / or the flow rate.

3. The method according to claim 1 or 2, wherein the input parameter set includes values ​​related to the thermal load of the receiving unit.

4. The method according to claim 1, wherein the input parameter set includes a value indicating time, such as a time of day.

5. The method according to claim 1, wherein the input parameter set includes values ​​indicating weather, such as outdoor temperature.

6. The method according to claim 5, wherein the system is provided with one or more further receiving units, each having individual supply lines and individual return lines connected to the primary supply line and the primary return line, respectively, and the temperature and / or flow rate of the thermal energy exchange medium in the individual supply lines and the individual return lines is known for the model.

7. The method according to claim 6, wherein the at least one predicted value further includes a value indicating the supply temperature of the thermal energy exchange medium in the individual supply lines of the further receiving unit.

8. The method according to claim 6, wherein the input parameter set includes at least one of the following: a value indicating the flow rate of the heat energy exchange medium in the individual supply lines and / or individual return lines of the further receiving unit; a value indicating the flow rate of the heat energy exchange medium in the primary supply line and / or primary return line; or a value indicating the return temperature in the individual supply lines and / or individual return lines of the further receiving unit.

9. The method according to claim 1, wherein the flow rate of the heat energy exchange medium in the primary supply line and / or the primary return line is calculated based on the value indicating the flow rate of the heat energy exchange medium in the single supply line and / or the single return line.

10. The method according to claim 1, wherein one or more previous predicted values ​​are provided to the input parameter set.

11. The method according to claim 6, wherein the value indicating the temperature of the heat energy exchange medium supplied by the heat energy supply unit via the primary supply line is measured indirectly using the measured temperature of the heat energy exchange medium in the individual supply lines of the further receiving unit.

12. The method according to claim 6, wherein the input parameter set includes values ​​indicating the thermal load of the further receiving unit, the flow rate in the further receiving unit, and / or the return temperature of the further receiving unit.

13. A system for characterizing the thermal energy distribution in a thermal energy exchange system using a data-driven model, wherein the thermal energy exchange system comprises a thermal energy supply unit configured to supply heating / cooling, and a plurality of receiving units configured to consume the heating / cooling supplied by the thermal energy supply unit, the thermal energy supply unit being connected to a primary supply line through which a thermal energy exchange medium flows from the thermal energy supply unit to the plurality of receiving units, the thermal energy supply unit being connected to a primary return line through which a thermal energy exchange medium flows from the plurality of receiving units to the thermal energy supply unit, and at least a subset of the plurality of receiving units, the primary supply line The connection to the supply line and the primary return line are modeled as a single supply line and a single return line, respectively, and the data-driven model is configured to operate under conditions where parameter values ​​indicating the temperature and flow rate of the thermal energy exchange medium in the modeled single supply line and the modeled single return line are not directly available and / or not monitored, and the system includes a processor configured to provide an input parameter set to a trained machine learning model system configured to output at least one predicted value, the at least one predicted value including a value indicating the characteristics of the thermal energy exchange system, and the input parameter set including at least a value indicating the temperature of the thermal energy exchange medium supplied by the thermal energy supply unit via the primary supply line.

14. A computer-aided method for controlling the supply temperature of a thermal energy exchange system, wherein at least one predicted value associated with the thermal energy exchange system is predicted using the method of claim 1, and the control of the supply temperature of the thermal energy exchange system is performed based on a predetermined target and the at least one predicted value.

15. A computer-aided method for controlling the thermal load of one or more further receiving units, wherein at least one predicted value associated with the thermal energy exchange system is predicted using the method of claim 6, and the control of the thermal load of the one or more further receiving units is performed based on a predetermined target and the at least one predicted value.

Images