JP7048667B2 - Methods and systems to control the energy supply to different units - Google Patents

Description

The present invention relates to a method of controlling energy supply to different units. Each unit is connected to multiple utilities (utilities) that receive energy to operate their energy system, and by at least one operating entity and / or by at least one utility, and / of a utility. Alternatively, a demand request signal is provided that requests a change in the demand for some form of energy. The present invention also relates to a corresponding system that controls energy supply to different units, preferably a system that implements the above method.

Methods and systems for controlling the supply of energy to different units, in which each unit is connected to multiple utilities that receive energy to operate its own energy system, are known in the art. In such methods and systems, at least one operating entity and / or at least one utility provides a demand request signal requesting a demand change of a utility and / or a form of energy. ..

Acronym list:
DR: Demand Response (Demand Response)
DRAS: Demand Response Automation Server
DER: Distributed Energy Resources
OpenADR: Open Automated Demand Response
RES: Renewable Energy Sources

As used herein, the terms utilities and energy networks are used interchangeably. The form of energy is, for example, electricity, gas or heat.

Due to the ongoing changes in energy systems caused by the high dissemination of RES and other types of DER, many utility providers and utilities are presenting DR programs as one of the possible means for energy management. There is. Units that join the DR program in the form of end users agree to change their consumption compared to normal use in situations where resources are scarce due to low supply or high demand. The types of rewards that users will get from utility providers to reduce their load are economical and the details are specified in the DR contract.

OpenADR is a standard designed to manage energy consumption through a communication model designed to send and receive DR signals between utility providers and power customers. OpenADR also shows how third parties interface with DRAS (Demand Response Automation Server), which is used to facilitate the automation of customer responses to various DR programs through the communication unit. It stipulates.

DR programs usually relate to the provision of electrical energy. But utilities are not the only ones who need to demand time-varying consumption. For example, a district heating operator may not be able to meet all demand, especially if its supply depends on an uncontrollable RES. Therefore, a system that considers DR for a plurality of utilities has appeared (see Patent Document 1 and Patent Document 2).

Patent Document 1 describes that a customer subscribes to a plurality of DR programs for different utilities. Forecasting future customer demand for each utility is based on CBL (Customer Baseline Load) calculation for each utility. Customer utility demand is predicted by establishing the CBL.

In Patent Document 2, a single master meter monitors consumption, resource generation costs, multiple utility types, and monitors and controls individual utility systems in the facility to enable utility cost adjustments. A multi-utility energy control system that determines and improves cost effectiveness is described. Consumption rates are monitored and compared to theoretical and / or historical data to identify unexpected changes in consumption, as well as peak demand, surges and sags. Also disclosed are software that control a utility consumption system by adjusting the current utility consumption in response to predetermined parameters.

As used in the following description herein, the term "cost function" is defined as follows.

In mathematical optimization, statistics, decision theory and machine learning, a loss function or cost function maps the value of an event or one or more variables to a real number that intuitively represents some "cost" associated with that event. It is a function. The optimization problem aims to minimize the loss function. The objective function is a loss function or its minus (sometimes called a reward function or utility function), and in the latter case, maximization is aimed at. In statistics, the loss function is usually used for parameter estimation, and the event in question is some function of the difference between the estimated and true values for an instance of the data (Source: Wikipedia http: //). en.wikipedia.org/). Note: The cost may be a real monetary cost, but it may also be a performance KPI degradation.

International Publication No. 2011/074925A2 U.S. Pat. No. 6,122,603

It is an object of the present invention to improve and further develop methods and systems for controlling energy supply to different units in order to enable highly efficient and reliable energy supply.

According to the present invention, the above object is achieved by the method with the configuration of claim 1 and the system with the configuration of claim 15.

As described in claim 1, the present method is characterized by the following. That is, the aggregator receives the demand request signal and assigns the requested demand change to the unit based on a process of negotiation with the unit to minimize the impact of the allocation on the future operation of other utilities. Run.

As stated in claim 15, the system is requested based on a process of negotiation with the unit to receive the demand request signal and minimize the impact of the allocation on the future operation of other utilities. It is characterized by having an aggregator that executes allocation of demand changes to units.

It has been recognized by the present invention that by using or providing an aggregator that receives a demand request signal and performs the allocation of the requested demand change to the unit, a very reliable and effective energy supply is possible. Become. Such allocations are based on a process of negotiation with the unit to minimize the impact of such allocations on the future operation of other utilities. Thus, the negotiation process takes into account the dependencies between related utilities and the impact of allocation on other utilities in order to provide balanced performance for that utility. This is an important feature. This is because the utility is not completely independent. For example, reducing the amount of electricity used to heat large buildings and sports facilities would affect the load on district heating utilities and vice versa. The impact of current changes in the consumption of one utility on other utilities can be considered by this method. As a result, highly efficient and reliable energy supply can be realized by using the resources of utilities in an optimized form.

In a preferred embodiment, the allocation may be performed dynamically. Such dynamic allocation responds very quickly to the utility's variable boundary conditions and / or environmental context parameters, providing a very sensitive and efficient allocation of requested demand changes. Such allocations may be made after a predetermined period of time. Preferably, the allocation may be triggered in real time, either continuously or by monitored changes in boundary conditions and / or environmental context parameters (eg, changes in meteorological conditions).

In a further preferred embodiment, the negotiation process may take into account the prediction and / or probability of one or more possible future demand change requests for the utility and / or other utilities. In view of such predictions, the allocation can take into account, for example, that a particular utility will not be able to provide a sufficient amount of energy within the expected future period. In that case, selecting the particular utility for ancillary energy supply within this future period would not be appropriate for a reliable and efficient energy supply. However, it is not only important to predict one or more possible future demand change requests for the utility, but also the probability of such possible future demand change requests. By considering such probabilities, a more reliable and effective energy supply to the unit can be obtained.

Predictions and / or probabilities may depend on different boundary conditions and / or environmental parameters. Preferably, the prediction and / or probability may depend on the weather, time, season and / or utility, unit or activity of the unit's energy system.

In a further preferred embodiment, the impact on future operation of the other utility may include the impact on one or more possible future demand change requirements of the other utility. In other words, the allocation may be performed taking into account the impact of one or more of the other utilities on possible future demand change requests. In this way, future possible demand change requests can be taken into account in the allocation.

For highly efficient allocation, the negotiation process may take into account feedback information and / or historical data and / or weather forecasts from at least one unit. Such a negotiation process allows consideration of different boundary conditions and environmental parameters.

Preferably, the feedback information from the unit may include an estimate of the maximum possible change or reduction or increase of the utility over the requested time period. Thus, the aggregator can use different information from different units regarding the maximum possible change range of the unit for the utility over the requested time period.

Even more preferably, the feedback information from the unit may include an estimate of how the maximum possible change or reduction or increase of the utility over the requested time period affects the load of other utilities to which the unit is connected. good. This information helps to allocate the requested demand changes very efficiently, with minimal impact on other utilities.

In a preferred embodiment, historical data may include a past allocation process per unit, including the amount and / or time series and / or duration of the allocated demand changes. Such historical data may help to reliably predict energy supply in future situations.

Alternatively or additionally, the negotiation process may be between different utilities for at least one given unit or each unit, preferably with respect to the assigned demand change and / or the size or amount of timely responsiveness. Suitable learned correlations may be considered. Such correlations provide a realistic prediction of the future situation.

The individual boundary conditions of the unit can be considered in the negotiation process. Preferably, the negotiation process may consider comfort levels and / or performance-specific priority levels and / or key performance indicator KPI requirements for activities or events, preferably for units.

The aggregator may notify the unit of the assigned changes for a very reliable energy supply. Such assigned changes are the result of a process of negotiation with the unit.

In a preferred embodiment, the aggregator may directly control the unit's energy system according to the assigned demand changes. There are various ways to implement the assigned demand change. Various communication systems can be used to control the unit's energy system.

Units are feasible by various entities. In a preferred embodiment, at least one unit may be a building. Multiple units may form a building campus.

The demand request signal may be provided by at least one utility. However, in a more preferred embodiment, the demand request signal may be provided by at least one operating entity. Such an operating entity may be one of the energy planning entities or units. Thus, for example, when a relocation within the unit is performed, the unit can initiate allocation with a demand request signal. This may be, for example, the addition of additional energy systems to the unit or the addition of additional rooms to the building.

Important aspects of the embodiments of the present invention are summarized as follows.

Embodiments of the invention may include dynamic distribution methods and systems for a multi-utility energy control aggregator between multiple independently operating energy management units. In this method and system
Energy change requests can be obtained from an external utility system and / or from the internal energy management system of the connected unit.
-The above energy change request is communicated between the aggregator and the connected operating unit, so that the unit can be united.
・ Negotiate the change tolerance for each utility and
-Providing information on applicable control or energy systems,
-Individual units allow aggregator activation for appropriate control or energy systems
・ The aggregator is
(1) Learned correlations between different utilities for each unit and connected units with respect to variables such as size, quantity, timely responsiveness, etc.
(2) History of change adaptation (size, time series, service level, responsiveness / capacity regarding duration) of each unit,
(3) Action-specific priority levels (eg, for activities / events, such as comfort levels, KPI requirements for scheduled operating conditions), and
(4) The control system or energy system according to an optimized distribution model that takes into account the prediction of future change requests (depending on the flexible context of weather, time, season, activity, etc.) for the unit group of the same or different utilities. It involves performing operations and predicting system performance with optional upcoming control signals, and calculating optimized distribution schemes between multiple autonomously operating units.

Preferred embodiments of methods and systems for multi-utility DR request distribution to units have the following features:

The method comprises the following steps.
1) Receive a DR signal requesting reduction for a given utility.
2) Communicate with the unit to inquire about the maximum possible reduction or increase and the resulting effect on other utilities.
3) Predict future DR reduction / increase demands and their probabilities.
4) Optimal allocation of current reductions or increases to minimize the impact on likely future demands.
5) Notify the units of the reduction / increase amount assigned to them.
6) Perform utility operation through the unit's utility control system.

By simultaneously managing DR signals for a plurality of utilities in a multi-site environment, the present invention can maximize the requirement fulfillment rate and minimize side effects for other utilities. This is achieved by considering probable future demands in the process of demand distribution.

An embodiment of the present invention proposes a system and a method for performing energy control management over a plurality of coordinating units. The system and method implement demand reduction / increase control through demand response request distribution that takes into account possible future DR demand predictions and the impact of current changes in the consumption of one utility on other utilities. The distribution of demand response requests is considered among the members of the unit group that subscribe to the demand response program and provide the opportunity to simultaneously service multiple utility programs such as electricity, hot water heating, gas and water. In order to achieve a high requirement fulfillment rate, this method considers the high probability of the most recent change request when negotiating and assigning a DR request to a unit.

Although we have considered aggregators and multi-utility systems above, it remains unsolved how load changes requested for one utility will have an impact on future demands for other utilities. This is an important issue. This is because the utility is not completely independent. For example, reducing the amount of electricity used to heat large buildings and sports facilities would affect the load on district heating utilities and vice versa. In the embodiment of the present invention, it is an object of the present invention to maximize the required achievement rate of the aggregator that distributes the DR signal to a plurality of utilities by considering the effect on the future DR signal that can be assigned at present. ..

In embodiments of the invention, the term "negotiation" preferably applies to exchanging information through communication protocols and mechanisms thereof (eg, OpenADR) defined in the art. This allows the aggregator and the unit to reach an agreement on how much and in what period the unit can meet its energy change requirements. Negotiations can consist of one or more communication exchanges. Note: In the broadest sense, instructing a unit to reduce energy by a certain amount over a period of time is also considered negotiation.

There are a number of possibilities for implementing the present invention in a preferred manner. For this, refer to the claims dependent on claim 1 on the one hand and the following description of the preferred embodiments of the invention exemplified by the drawings on the other hand. When a preferred embodiment of the present invention is described with reference to the drawings, a general preferred embodiment according to the teaching of the present invention and examples thereof will be described.

It is a figure which shows typically the embodiment of the system which controls the energy supply to different units by the DR request distribution by this invention. It is a figure explaining the embodiment of the method of performing a change request distribution by this invention. It is a figure which shows typically the operation execution of the optimization decision of the aggregator in the embodiment of different enforcement. It is a figure which shows typically the possible communication between an aggregator server and a unit by this invention.

FIG. 1 schematically illustrates an embodiment of a system for controlling energy supply to different units according to the present invention. Figure 1 shows an aggregator servicing energy change requests (eg, through a DR signal, through an on-campus scheduling unit) distributed under a set of independent energy control units that are flexible with respect to multiple (interrelated) energy dimensions. Shows a system for multi-utility energy control management functions housed through. Change requests arrive at the aggregator over m different utilities, preferably through the communication network, preferably the Internet. The end customer (here the unit) has a network connection to the above utilities. The aggregator determines how to allocate the required amount of reduction / increase based on additional information such as feedback, historical data, and weather forecasts received from the unit. The main goal of the energy control management function in the aggregator is
(1) History of unit change adaptation (eg size, time series, service level),
(2) Forecasting future change requests (preferably depending on weather, time, season) for units of the same or different utilities,
(3) Learning unit-specific multi-utility correlations for size and quantity, and (4) Behavior-specific priority levels (eg, KPI requirements for comfort levels).
It is to achieve the received change request at a high rate and realize a dynamic distribution method among a plurality of units.

The present invention predicts future energy change demands while distributing the required reduction / increase of the utility, minimizing the effect of current change allocation on utilities that may require changes in the near future. Suggest to try. The main idea of the proposed method is shown in FIG.

In the following description, an example of the present invention relating only to the consumption change request realized by the DR signal of the present major interest will be described. Other sources of change requests include, for example, energy planning entities in campus energy networks.

The aggregator receives a DR request that specifies the utility k and the period during which the utility k should be reduced, along with the desired reduction amount. In response to this DR signal, the aggregator estimates for each unit the maximum possible reduction in utility k over a specified time period and how this reduction affects the load on other utilities to which the unit is connected. Request an estimate of what to do. This feedback is provided by a DR module that can model how reductions in one utility affect other utilities. The aggregator also predicts the future DR that may occur over the period associated with the current requirement and the probability of these events. This is done based on historical data and weather forecasts. With all this data in place, in the next step, the aggregator will assign the reduction in utility k to the unit and minimize its effect on utilities that are likely to be subject to (reduction) demands in the near future. And. This requirement is translated by the present invention into the corresponding control command to be enforced. In the final step, the aggregator notifies the unit of the requested reduction for the unit and the control enforcement of the unit's utility control system is performed.

The communication process that occurs between the aggregator and the unit is shown in FIG. 4 step by step. In the first step, the aggregator sends the specified aggregated DR request that arrived to all units. This designation includes the utility (k) that needs to be reduced, the period for which reduction is required (given from _start to _tend ), and the desired reduction amount A. Each unit ⁱ returns the maximum possible reduction in utility _k ( _{uk_max} ⁱ ) and an estimate of the effect of this reduction on all other utilities (given as u ₁ ⁱ , ..., u mi). do. In the final step, the aggregator sends its assigned reduction amount α _i * _{uk_max} ⁱ (where α _i ∈ (0, 1)) to each unit i.

In one preferred embodiment, each unit is connected to different controllable systems associated with different utilities. The control module of each unit interacts with the aggregator to obtain management information [ctrl (α _i * _{uk_max} ⁱ )] in the DELEGATE ENFORCED mode and directly through the communication protocol specified in the art. Alternatively, the controllable system is activated indirectly through an intermediate interconnected system [act (α _i * _{uk_max} ⁱ )].

In another preferred embodiment, the controllable system is directly connected to the aggregator. That is, the aggregator is allowed to control the controllable system through communication protocols defined in the art. In the case of the present embodiment, the aggregator control unit may send management information [ctrl (α _i * _{uk_max} ⁱ )] to the control module of the unit (or may omit this) in order to notify the operation. ) Operates the controllable system directly from the aggregator server in DIRECT ENFORCED mode by [act (α _i * _{uk_max} ⁱ )], or by remote enforcement (REMOTE ENFORCED) to the location of the unit. Operated through a remotely housed aggregator control unit. See Figure 3 for details.

In a modification of the invention, the unit may initiate communication with the aggregator, for example informing the aggregator of the addition or reduction of required energy over a specified time range. This information may be derived from past energy usage statistics, energy forecasts, etc. The aggregator then uses this information to request the other units connected to the aggregator to make appropriate energy changes, as defined in the present invention.

Specific Embodiment in the case of DR for a plurality of buildings If there are n large-scale buildings B ₁ , ..., B _n that are subscribed to the DR program of m utilities and are linked by the same aggregator, the utility k After receiving a signal that reduces the amount by the amount of the period (t _start , _{tend), the aggregator will determine the maximum possible reduction amount for each building (expressed as uk_max i} ^for the building i) and this reduction. ^Inquire about the costs u ₁ ⁱ , ..., u _mi that result from other utilities. The resulting cost u _j ⁱ is an estimate of the building i to what extent reducing utility k by _{uk_max} ⁱ affects utility j. The value of _ujii can be positive or negative depending on the effect caused by the reduction in utility ^k . If a decrease in utility k leads to an increase in utility j, the value will be positive. However, the opposite is also possible if utilities k and j are coupled together as input to a process that can be replaced by a third utility.

In the next step, the aggregator needs to predict future DR events based on historical data and possibly weather forecasts. This prediction is limited to the period during which the utility is affected by the reduction in utility k over a period (t _start , _tend ). In addition to possible events, the aggregator should also predict its probability. Bayesian networks can be used for this type of prediction. In this way, a set of probable events and their probabilities are obtained (E = {( _{e 1} , p ₁ ), ..., (ej, p _j )}). With a certain threshold p _border , events with a higher probability than the p _border can be considered as likely. H represents a set of all utility indexes h such that the probability _ph of the event e _h requiring reduction in the utility _h is at least a loader.

At this stage, the aggregator needs to solve the following optimization problem in order to assign the reduction in utility k to the building. The control variables are α ₁ , ..., α _n , and the ratios of the maximum possible reduction amounts _{uk_max} ¹ , ..., _{uk_max} ⁿ required for the buildings B ₁ , ..., B _n , respectively. show. Any standard optimization method, such as a genetic algorithm or pseudo-annealing, can be used to solve this problem. In the case of this specific embodiment, since the problem formulation is linear, linear programming can be used to solve it.

Problem formulation:
Objective function:

として推定される。ただしα_ｉは制御変数（建物Ｂ_ｉに対して要求される低減量α_ｉ＊ｕ_{ｋ＿ｍａｘ} ^ｉを決定する）であり、ｕ_ｈ ^ｉは、建物Ｂ_ｉにおけるユーティリティｋの低減ｕ_{ｋ＿ｍａｘ} ^ｉに対するユーティリティｕ_ｈにおけるコストである。この関数は、ユーティリティｉにおけるコストが、ユーティリティｋの低減とともに線型に増減すると仮定してコストの推定を行う。 The purpose is to minimize the impact of the requested reduction on utilities (h ∈ H) that are likely to be required to be reduced in the near future. _wh is a weighting factor used to give different importance to different utilities. The weighting factor should be selected according to the specific application case, but one option is to use the event probability _ph as a measure of relative importance. The cost for utility h is

Estimated as. However, α _i is a control variable (determines the reduction amount α _i * u _{k_max i} required for the building Bi ⁾ , and u _h ⁱ is the utility ^u for the reduction u _{k_max i} of the utility k in the building Bi. It is a cost in _h . This function estimates the cost on the assumption that the cost in the utility i increases and decreases linearly as the utility k decreases.

ただし、εは、要求された低減量からの許容されるずれを表す。要求された低減が達成不可能な場合によりよく振る舞うもう１つの可能なアプローチは、この絶対値に、重み因子付きで目的関数を加算することである。制約を用いたアプローチは、問題が実行不可能な場合に、要求される値を低くした問題の再実行を必要とする。 Constraints:
Attempts should be made to allocate the entire desired reduction amount as follows:

Where ε represents the permissible deviation from the requested reduction amount. Another possible approach that behaves better when the required reduction is unattainable is to add the objective function to this absolute value with a weighting factor. The constraint approach requires re-execution of the problem with a lower required value if the problem is infeasible.

There are two more constraints that require control variables to belong to the range [0,1].
1) For all i α _i ≧ 0
2) For all i α _i ≤ 1

After the problem is solved, the unit in charge of each building ^Bi is notified of the requested reduction amount α _i * _{uk_max i} and controls the building management system accordingly. In a modification of this embodiment, the aggregator is allowed to directly control the building management system after solving the optimization problem. In that case, it is optional to notify the reduction request for each individual unit.

Based on the above description and description of the accompanying drawings, one of ordinary skill in the art will be able to conceive of many modifications and other embodiments of the invention. Therefore, the present invention is not limited to the disclosed specific embodiments, and variations and other embodiments should be understood to be included in the appended claims. Although specific terms are used herein, they are used only in a generic and descriptive sense and are not intended to be limiting.

Claims (15)

In a way of controlling the energy supply to different units, each unit is connected to multiple utilities that receive energy to operate its own energy system, by at least one operating entity, and / or at least one. The utility provides a demand request signal requesting a demand change for the first utility of the plurality of utilities.
The aggregator receives the demand request signal and
The predictive means requests each unit to estimate the impact of the allocation of the requested demand change to each unit, receives the estimated impact from each unit in response to the request, and each of the demand changes of the first utility. Predict the effect of the allocation to the unit on the future operation of the second utility, which is a utility other than the first utility among the plurality of utilities.
The optimization means performs the allocation of the requested demand change to the unit so as to minimize the predicted impact of the allocation.
A method of controlling the energy supply to different units. The method of claim 1, wherein the allocation is dynamically performed in real time. The optimization means according to claim 1 or 2, wherein the optimization means considers the prediction and / or probability of one or more possible future demand change requests of the first utility and / or the second utility. the method of. The method of claim 3, wherein the prediction and / or probability depends on the weather, time, season and / or activity. The invention according to any one of claims 1 to 4, wherein the influence on the future operation of the second utility includes the influence on one or more possible future demand change requests of the second utility. Method. The method of any one of claims 1-5, wherein the optimization means takes into account feedback information and / or historical data and / or weather forecasts from at least one unit. The method of claim 6, wherein the feedback information from the unit comprises the maximum possible change of the utility over the requested time period, i.e., an estimate of reduction or increase. The feedback information from the unit is characterized by including an estimate of how the maximum possible change or reduction or increase of the utility over the requested time period affects the load of the second utility to which the unit is connected. The method according to claim 6 or 7. 13. the method of. The optimization means is characterized by taking into account well-learned correlations between different utilities for at least one predetermined unit or each unit with respect to demand change and / or timely responsiveness size or quantity. The method according to any one of claims 1 to 9. Claims 1-10, wherein the optimization means takes into account comfort levels and / or performance-specific priority levels and / or key performance indicator KPI requirements for activities or events. The method according to any one of the above. The method of any one of claims 1 to 11, wherein the aggregator notifies the unit of the assigned changes. The method of any one of claims 1-12, wherein the aggregator directly controls the energy system of the unit according to the assigned demand change. The method according to any one of claims 1 to 13, wherein at least one unit is a building and / or the at least one operating entity is an energy planning entity or unit. In a system that controls energy supply to different units and implements the method according to any one of claims 1 to 14.
Each unit is connected to multiple utilities that receive energy to operate their energy system, and by at least one operating entity and / or by at least one utility, the demand for the first utility of the plurality of utilities. A demand request signal requesting a change is provided,
The aggregator,
The means for receiving the demand request signal and
Requests each unit to estimate the impact of the requested demand change allocation to each unit, receives the impact estimation from each unit in response to the request, and assigns the demand change of the first utility to each unit. Is a predictive means for predicting the influence on the future operation of the second utility, which is a utility other than the first utility among the plurality of utilities.
An optimization means for performing the allocation of the requested demand change to the unit so as to minimize the predicted impact of the allocation .
A system that controls the energy supply to different units.

Images