WO2021121881A1

WO2021121881A1 - Determination of the structure of an energy system that is resilient to a failure of one or more of the components thereof

Info

Publication number: WO2021121881A1
Application number: PCT/EP2020/083204
Authority: WO
Inventors: Lukas HÖTTECKE; Sebastian THIEM
Original assignee: Siemens Aktiengesellschaft
Priority date: 2019-12-20
Filing date: 2020-11-24
Publication date: 2021-06-24
Also published as: DE102019220347A1

Abstract

The invention relates to a computer-assisted method for the numerical determination of the structure of an energy system (1) that is resilient to a failure of one or more of the components (21,..., 23) thereof, wherein the energy system (1) is described by a target function intended for optimization such that the structure of the energy system (1) is determined by calculating an extremum of the target function. The method is characterized in that the extremum is calculated under the secondary condition that a critical load P _L* (42) is further covered by one or more contributing components i (21,..., 23) of the energy system (1) for a call-off duration T _D in such a manner that components and/or processes dependent on the critical load P _L* (42) are not substantially impaired, wherein, to this end, the periodic change of the capacity P _i* of each contributing component (21,...,23) within a response time T _R of the energy system (1) is limited by the nominal capacity P _net,i of the component (21,..., 23) in question weighted by the starting time T _net,i. The invention further relates to an associated energy management system and an energy system (1).

Description

description

Determination of the structure of an energy system that is resilient to the failure of one or more of its components

The invention relates to a method according to the preamble of claim 1, an energy management system according to the preamble of claim 12 and an energy system according to the preamble of claim 14.

Energy systems typically include several components, in particular energy-related systems, for example energy conversion systems, consumption systems and / or storage systems. Multimodal energy systems are energy systems that generate and / or provide energy in several forms of energy. In particular, a multimodal energy system for an energy consumer, for example a building, an industrial plant or a private plant, provides one or more forms of energy, the provision in particular by converting different forms of energy, by transporting different forms of energy and / or by means of stored forms of energy he follows. In other words, the various forms of energy, for example heat, cold or electrical energy, are coupled by means of the multimodal energy system with regard to their generation, their provision and / or their storage.

A typical industrial multimodal energy system, for example in the food industry, has several energy technology systems (technologies for energy conversion), various storage devices (storage technologies), a connection to a (public) electricity network, a connection to a (public) gas network and / or hydrogen network and several electrical and thermal loads.

It is known to design energy systems as efficiently as possible by means of an energy system de- sign process (energy system analysis) to design, that is, to determine their optimal structure. For this purpose, a mathematical model of the energy system is used, which enables the energy system to be optimized with regard to a target function by means of a numerical optimization method. Typically, a plurality of physical parameters (input parameters), for example predictions of load profiles and / or state measurements, for parameterizing the mathematical model or the target function, as well as several secondary conditions, are required for this. In other words, energy systems, in particular multimodal energy systems, are designed using data-based optimization methods.

Such an energy system design method is known, for example, from the document EP 3561743 A1.

An energy system design method thus determines the most optimal structure of the energy system, for example with regard to its energetic efficiency. The structure of the energy system is formed in particular by its components, the dimensions and / or capacities of its components, the costs of its components, energy flows, power flows and / or by its energy transfer lines.

For example, a combined heat and power plant (CHP plant) provides electricity, i.e. electrical energy, during high load times. The waste heat from the CHP plant can in turn be at least partially temporarily stored by means of a heat storage device and used to cover thermal loads. The optimization can be based on measurement data relating to the electrical and thermal loads mentioned.

However, in known methods, measurement data relating to the (external) power grid are not used. This jeopardizes the security of supply of the energy system. So-called backup generators are used to prevent this. As a result, a certain resilience of the energy system, which is not optimal, can be provided. Typically, energy systems that have a certain resilience, for example to a failure of the external power grid, are overdimensioned.

Here, the term resilience denotes, for example, the ability of the energy system not to fail completely in the event of faults, failures and / or partial failures and / or disruptive events, but rather to maintain critical loads, i.e. to cover them. In this sense, a load is critical if it has to be maintained at least partially, in particular completely, even when a fault, a failure and / or partial failure and / or a disruptive event occurs.

A typical malfunction or a typical failure is a malfunction or a failure of an external power grid connected to the energy system but with respect to this. Failures or malfunctions can therefore be triggered by internal and / or external events in relation to the energy system. In particular, a disruptive event is a power failure, a hacker attack on a control room and / or a technical failure of CHP systems.

Failure of the external power grid is the most common cause of electrical supply interruptions. Long-term interruptions (network interruptions) can be avoided through redundancies and switchovers within the external power grid. Short-term power interruptions (<3 minutes) are typically caused by switchovers and correlate with atmospheric events or effects, such as thunderstorms. Thus, in the event of short-term supply interruptions, there are hardly any opportunities for intervention by a network operator of the electricity network. According to EN 50160, up to several hundred short supply interruptions (<3 minutes) per year can therefore be expected. Electrical supply interruptions can lead to considerable damage, for example through damage to production facilities or through defective products. Newer production systems, especially with regard to the digitization of industrial energy systems or production (Industry 4.0), are particularly vulnerable due to their sensitive control electronics (control unit). The failure of critical processes can also pose a security risk to people.

It is therefore imperative to provide energy systems that are resilient and yet as efficient as possible.

In this regard, from the document WO 2019/110220 A1 a method for providing the best possible resilience of an energy system is known.

The present invention is based on the object of providing a method for determining the structure of an energy system which leads to an improved resilience of the energy system. Furthermore, an energy system with improved resilience is to be provided.

The object is achieved by a method with the features of independent claim 1, by an energy management method with the features of independent claim 13 and by an energy system with the features of independent claim 15. In the dependent Pa tent claims, advantageous configurations and further developments of the invention are given.

The computer-aided method according to the invention for numerical determination of the structure of an energy system that is resilient to the failure of one or more of its components, the energy system being described by a target function provided for optimization, so that the structure of the energy system is determined by calculating an extremum of the target function is characterized by that the calculation of the extremum takes place under the secondary condition that a critical load P is covered by one or more contributing components i of the energy system for a call duration r _{D in} such a way that components and / or processes dependent on the critical load P are essentially not, in particular not, are impaired, whereby the time change in the power P _j ^{* of} each contributing component within a response time T _{R of} the energy system is limited by the nominal power P _{net, i of} the respective component weighted with the start-up time T _{net i} becomes.

There can be one or more critical loads.

Services during or after the failure are marked with an asterisk. The services at or before the failure show no asterisk.

As described above, a failure is any possible disruption in the operation of the energy system, in particular the at least partial failure of a component of the energy system and / or the at least partial failure of an external power grid with respect to the energy system to which the energy system is connected for exchanging electrical energy ( Reference and / or feed) is connected. If this power grid fails, certain components of the energy system also fail. Other networks, data networks and / or supply networks, for example gas networks and / or hydrogen networks, can be provided as an alternative or in addition.

The component-specific variable, such as the performance P _j ^* during or after the failure and / or the performance P _j during or before the failure and / or the performance P _L or P _L are fundamentally time-dependent, _{i.e. formally P j} ^{* applies} = Pi ^* (t) or P _j = P _j (t) and / or P _L = P _L (t) and / or P _L = P _L (P). Furthermore, the retrieval duration P _{D can} also be time-dependent, that is to say, for example, as well as the other variables mentioned, it depends on the time of the failure. However, the time dependency becomes simple notation is not explicitly written out here.

By determining the structure of the energy system (design of the energy system), in particular its components, the dimensions and / or capacities of its components, the costs of its components, its energy flows, its power flows and / or its energy transfer lines and / or the energetic couplings determined between its components. Through the optimization, the structure of the energy system and its energy flows or power flows is mapped or modeled holistically so that synergy potentials between the components and / or the various forms of energy can be determined. In particular, the energy-related structure of the energy system is determined.

A load is critical if it must be at least partially, in particular completely, maintained, that is to say must be covered, even when the failure occurs. In this case, the load must be at least partially covered if dependent components and / or processes are significantly impaired by a disturbance in the load. An impairment is essential, for example, if an operation of the component that is dependent on the critical load is no longer possible and / or the process that is dependent on the critical load can no longer be carried out or maintained. In other words, the critical load continues to be covered during the failure in such a way that critical processes associated therewith are not inadmissibly impaired.

An optimization within the meaning of the present invention is a method for minimizing or maximizing (calculating an extremum) of an objective function. The minimization or maximization of the objective function is typically extremely complex and can therefore only be done numerically. The objective function characterizes a property or a size of the system, for example the carbon dioxide emissions or the operating costs of an energy system. The objective function is thus a mathematical model of the energy system. The objective function has parameters and variables. The result of the optimization, i.e. in the present case the structure of the energy system, corresponds to the values of the variables of the objective function. The parameters are typically load profiles based on measurement data and parameterize the target function specific to the energy system. Furthermore, the optimization is typically carried out taking into account several secondary conditions, for example energy conservation.

Constraints, constraints or constraints - here collectively referred to as constraints - are conditions, properties and / or relations that the parameters and / or variables of the optimization process must meet. These can be given as an equation and / or inequality and / or explicitly describe a set of permissible values of the parameters and / or permissible values of the variables.

For example, the objective function characterizes the total carbon dioxide emissions of the energy system. The total carbon dioxide emission comprises the individual carbon dioxide emissions of the components of the energy system. Each of the components can provide a specific electrical and / or thermal output for a specific amount of carbon dioxide. Thus, the electrical and / or thermal outputs are the variables of the objective function and the amounts of carbon dioxide per generated electrical and / or thermal output (specific amount of carbon dioxide) are the parameters of the objective function. The optimization minimizes the total carbon dioxide emission (target function). As a result, the energy system can be designed and / or operated as optimally as possible with regard to its carbon dioxide emissions.

According to the present invention, the power P _j ^{* of} each component contributing to meeting the critical load becomes within a response time T _{R of} the energy system by the nominal power P _{net, £} of the respective component (from above) weighted with the start-up time T _neti. This takes place within or for a retrieval period T _O. The method thus has three characteristic types of time or duration, namely the response time T _R , the retrieval duration 7j ₎ and the individual start-up _{times T neti} .

The response time T _R is the duration of the change in operation of the energy system that occurs due to the failure. In other words, this is the period of time in which the energy system can react, or reacts, to the failure. It can therefore also be referred to as the response time. Typically, each component of the energy system has a different response time. The response time T _R can be defined as the longest of the response times of the components. In other words, the response time T _R of the energy system is determined by the component of the energy system that is slowest in terms of the reaction time. The response time T _{R of} the energy system is thus determined by its components. In particular, CHP systems and / or emergency power systems cannot adjust their operating point at will.

The call duration T _O is the time in which the supply of the critical load must be ensured. In other words, the state that results from the change in the operating points or working points must be maintained for the period T _O. The retrieval time T _O is typically longer, in particular significantly longer, than the response time T _R. The retrieval duration T _O is determined, for example, and can correspond to an average expected downtime. For example, the call duration 7j _{) is determined} on the basis of local availability statistics of a power grid that is external to the energy system. If such data are not available and / or are not available, empirical values from comparable locations can be used. Thus, the inventive Process can be optimally adapted to the quality of supply at the location of the energy system.

The start-up time T _{net i} is the component-specific period of time in which the respective component increases its output from zero to the nominal output P _{net, £} (from 0 percent to 100 percent). The respective start-up time T _{net i} is thus determined by the respective associated component.

The change in the power of a component is limited during the call time by the nominal power P _ne t, i of the component (from above) _{weighted with the start-up time T net i.} In ^¬ play, the change of the performance of the component is less than or equal to P _n is Et, i / _n T et, i (operating point in linear change of Be ^¬) for all time points within the Äbrufdauer _7j). The restriction applies to all contributing components, i.e. for all components that contribute to covering the critical load. This ensures that the change in operating status of the respective contributing component is taken into account. In other words, each contributing component needs a certain amount of time to change the operating state, for example to increase its performance. However, the performance cannot be increased faster than P _n et, i / 7net, i. This is taken into account by the secondary condition according to the invention. There is thus a transition phase, which _{corresponds to the response time T R} , in which the required stationary power to cover the critical load is not immediately available. In other words, according to the invention, the start-up time or the start-up times of the components is taken into account in the optimization.

A further advantage of the present invention is that, in particular, short-term interruptions in the external power network are compensated for and the operation of the energy system is therefore not significantly impaired. The present method thus enables critical loads and / or processes and / or systems to be safeguarded at any point in time. Here- downtime costs can be minimized and at best avoided completely.

Another advantage of the present invention is that it enables the energy system to operate in a way that is beneficial to the grid. In particular, this enables isolated operation of the energy system, in which decentralized energy-technical systems, for example generating systems and / or storage systems, can be operated independently of the power grid or the public power grid. As a result, the energy system can provide a flexibility which contributes to the stabilization of the overall network, in particular the electricity network, or can be used for accelerated network reconstruction.

Furthermore, for many applications and / or integration of the energy system, in particular for integration in a local energy market, the highest possible temporal resolution is required, for example every minute. The method according to the invention correctly maps energy management-relevant bills to a low temporal resolution, for example to 15-minute mean values.

The energy management system according to the invention comprises at least one control unit for controlling several components of an energy system, the control of the components being based on an objective function provided for optimization, so that control values for controlling the components are determined by calculating an extreme of the objective function. The energy management system according to the invention is characterized in that the control unit is designed in such a way that the calculation of the extremum takes place under the secondary condition that a critical load is covered by one or more contributing components i of the energy system for a retrieval period r _{D in} such a way that that components and / or processes that are dependent on the critical load P are essentially not, in particular not, impaired, with the time change in the power P _j ^* each contributing to this. lowing component within a response time T _{R of} the energy system is limited by the nominal power P _ne t, i of the respective component weighted with the start-up time T _neti.

Similar and equivalent advantages and / or configurations result for the method according to the invention.

The energy management system particularly preferably has a data interface which enables data to be exchanged between the energy management system and a local energy market, a cloud server and / or a decentralized data center.

The energy system according to the invention is characterized in that it comprises an energy management system according to the present invention and / or one of its design.

Similar and equivalent advantages and / or configurations result for the method according to the invention and / or for the energy management system according to the invention.

It is preferred here if the call duration T _{O of} the associated energy management system is determined as a function of the location of the energy system.

As a result, a local supply quality can advantageously be taken into account, in particular with regard to a power grid to which the energy system is connected.

Furthermore, the energy system can comprise spatially distributed dividing systems. This is the case, for example, with several production sites.

According to a preferred embodiment of the invention, the change over time in the powers P _j ^{* of} each contributing _{component is} limited _{by P net, i} / T neti. In other words, there is advantageously a linear change in the operating point of the respective component. As an alternative or in addition, there is a non-linear change in the power for at least one of the contributing components. For example, for a non-linear change, the temporal performance of at least one of the contributing components is _{limited by Gi (P net i} / T _{net i} ), where G _{j describes} the non-linear dependence of component i. Alternatively or additionally, Gi is independent of the component i, that is, the nonlinear dependence is given for all i by G (P _{net i} / T _{net i} ). In the linear case, for example, G (x) = x with x = iet, i / ^ net, i

In an advantageous development of the invention, the power P _j ^{* of} each contributing component within the response time T _{R of} the energy system (1) is _{limited by Pi + P n} e _{t, i} T _R / T _{net i} , where Pi is the power of the respective Component i referred to before or on failure.

Essentially, Pi + P _n e _{t, i} T _R / T _{net i is} the integral of the condition that the temporal change in the performance of the respective component by P _n et _, i / T _n et _{, i} within the response time T _R of is limited above. The two conditions mentioned are therefore equivalent. In other words, the condition means that only part of the respective nominal power can be provided by the respective component in order to continue to supply the critical load within the response time. Thus, the condition Pi + P _n e _{t, i} T _R / T _{net i} <P _{net i} must also always be fulfilled. In other words, the respective performance of the component within the response time is limited by Pi + P _n e _{t, i} T _R / T _{net i} and by P _{net, £} from above.

In other words, it is preferred if the power P _j ^{* of} each contributing component is limited within the response time T _R by its respective nominal power P _{net, £} (from above). According to a preferred embodiment of the invention, non-critical loads are not covered during the retrieval period T _O.

In other words, only critical loads or the critical load are advantageously supplied further. As a result, there is advantageously basically more power available for supplying the critical load or the critical load.

In an advantageous development of the invention, at least one of the contributing components is an energy storage device, with the calculation of the extremum taking place under the further secondary condition that the power P _ES provided by the energy storage device during the call period T _O satisfies the inequality? 7 _ES P _ES r _D <SOC Cap _ES fulfilled, where h _{E $} denotes the discharge efficiency of the energy store, SOC denotes the state of charge of the energy store before or in the event of failure, and Cap _ES denotes the capacity of the energy store.

In other words, the energy store can only provide part of its nominal output (total output) (for a certain time). This physical limit is integrated into the optimization by the secondary condition t] _ES P _ES T _O <SOC Cap _ES , so that advantageously the optimization or the solution to the optimization (structure of the energy system) takes into account the physical boundary conditions of the energy store. The variables mentioned, in particular the state of charge SOC, can be time-dependent, that is, SOC = SOC (t), for example, so that they can also depend on the time of the failure.

The energy store advantageously provides flexibility within the energy system. Alternatively or in addition, the energy storage system can reduce peak loads if an energy management system is in place. The energy store is preferably an electrochemical energy Storage, in particular a battery storage. One or more energy stores can be provided.

According to a preferred embodiment of the invention, the calculation of the extremum takes place under the secondary condition _{during the retrieval period T O}

for all contributing components i.

This advantageously ensures that the critical load is covered by the contributing components during the failure, that is to say within the retrieval period r _D.

In an advantageous development of the invention, the critical load during the call duration T _{O is} determined by P = P (P _L ), where P _L denotes the power of the critical load before or during failure, and F the dependence of the critical load P on _{L describes} dependent components and / or processes.

In other words, the function F models the condition that components and / or processes dependent on the critical load, that is to say critical processes associated with the critical load, are essentially not impaired, that is to say are not impaired in an impermissible manner. In particular, short-term and unforeseeable load peaks must also be covered. For example, in order to maintain a critical process, a certain percentage a of the critical load present before the failure could additionally be made available by the components of the energy system. In other words, for example, P = P (P _L ) = (1+ <^) P _{L /} with non-linear functions F being provided as an alternative or in addition.

According to an advantageous embodiment of the invention, an energy store, a combined heat and power plant and / or an emergency power system are used as contributing components. The components mentioned can advantageously be operated dynamically with regard to the provision of power. In particular, the energy store and / or the backup system can respond essentially immediately to a failure, so that the response time of the energy system is thereby advantageously reduced overall.

In a preferred development of the invention, the respective structure of the energy system is determined for a plurality of retrieval periods T _O.

In other words, several structures of the energy system are ascertained or determined for several associated retrieval periods. The structure of the energy system can thereby advantageously be adapted _{to the call duration T O.} The retrieval duration can depend on the location of the energy system, in particular on the region and / or country in which the energy system is installed or is to be installed. The retrieval duration can therefore be significantly different for different locations or intended locations of the energy system. If the structure is determined as a function of the retrieval period, a structure of the energy system that is as optimal as possible for the location or region can be determined. In other words, the location of the energy system is thereby advantageously also taken into account. _7j) for is particularly preferably determined from local availability statistics of a local power grid that is superordinate with respect to the energy system, and the structure of the energy system is determined _{for this determined call duration T O.} The energy system is connected to the aforementioned power grid.

According to a preferred embodiment of the invention, the respective value of the objective function and a respective value of a failure function are determined for the several specific structures, the failure function quantifying the failure risk of the critical load as a _{function of the call duration 7j), and a structure from the specific structures} he with- which has the smallest value of the sum formed from the objective function and failure function.

The value of the objective function (extremum) characterizes the most optimal structure of the energy system. In other words, the value of the objective function, i.e. its extremum, is determined as a function of the call duration. The objective function typically characterizes a variable that should be minimized as much as possible, for example the carbon dioxide emissions of the energy system, its primary energy use and / or its costs (operating costs and investment costs). The objective function can, however, be advantageously expanded to determine the best possible structure with regard to a failure and the creation of resilience, since this typically assumes that the energy system is operating without such a failure, that is, from the best possible operation. However, a failure is associated with a certain risk of failure, which is described by the failure function. The failure function is to be added to the original target function, so that a new target function that takes into account the case arises depending on the call duration. The optimal structure now results from the minimum of the new objective function mentioned. In other words, the optimal structure is the solution with the lowest value of the sum formed from the target function, for example the lowest total of the costs (English: Total Expenditure; Totex), and the lowest default risk, for example a financial default risk.

Further advantages, features and details of the invention he will play from the Ausführungsbei described below and based on the drawings. The following are shown schematically:

FIG. 1 shows an energy system according to an embodiment of the present invention;

FIG. 2 a first diagram; and Figure 3 shows a second diagram.

Identical, equivalent or identically acting elements can be provided with the same reference symbols in one of the figures or in the figures.

FIG. 1 shows an energy system 1 according to one embodiment of the present invention. Here, the configuration is described below using an electrical supply system at a node. Alternatively or in addition, further energy-related nodes or networks and / or supply networks, for example with regard to heat, cold and / or chemical substances, for example gas, can be provided.

The energy system 1 comprises several energy-technical compo nents 21, ..., 23, in this case an energy storage device 21, for example a battery storage, a combined heat and power plant (CHP plant) 22, for example a block-type thermal power station and an emergency power system 23, for example a Diesel generator. The energy system 1 also includes a control unit 20.

The energy system 1 is connected to a power grid 3 that is external to the energy system 1.

Furthermore, the energy system 1 comprises a first load 41 and a second load 42. Alternatively or in addition, external loads that are supplied by the energy system 1 are also conceivable.

In the present case, the first load 41 is a non-critical load, which means that it does not necessarily have to continue to be covered in the event of a failure of the power grid 1. In contrast to the first load 41, the second load 42 is a critical load which must at least partially be covered in the event of a failure of the power grid 3. In principle, all loads can be used as Non-critical load or critical load who categorized, whereby a load - depending on the time and / or type of failure - can be a non-critical or a critical load.

The energy-related components 21,..., 23, the control unit 20 and the critical load 42 form a protected system 2 within the energy system 1. This protected system must supply itself with power / energy during the failure of the power grid 3 and maintain certain loads, in the present case the critical load 42. The first load 41, on the other hand, as a non-critical load, is not part of the protected system 2, that is to say that it is no longer supplied in the event of a failure. If the power grid 3 fails, the protected system 2 forms an island grid in this sense.

In the event of a failure of the power grid 3, the critical load 42 must continue to be covered for a retrieval period. The call duration can correspond to a typical downtime of the power grid 3 and / or be at least as long as the downtime of the power grid 3. The energy-related components 21,..., 23 all contribute to covering the critical load 42 in the present exemplary embodiment. They are therefore contributing components.

Typically, the energy system 1 cannot react immediately to a failure of the power grid 3. The energy system 1 requires a certain response time in order to adjust the performance of the respective contributing components 21, ..., 23 to the required new operating state. For example, the CHP system 22 and / or the emergency power system 23 must be started up. The response time of the energy system 1 is then determined, for example, by the longest component-specific response time. According to the present invention, the response time is taken into account when controlling (the concept of controlling includes the concept of regulating) the energy system 1 by the control unit 20. This is done by that the temporal change in the power of the respective load-bearing component 21, ..., 23 is _{limited by P n} et, i / T _n et, i (in the linear case), where P _{netj is} the respective nominal power and T _{net i is} the start-up time the respective component i denotes. The mentioned restriction is taken into account as a secondary condition in an optimization method which is used to determine the most optimal structure of the energy system 1 or the most optimal possible operation of the energy system 1. During operation, the control unit 20 performs the optimization, for example as part of an energy management system. In this sense, the control unit 20 forms a model-predictive control or this is designed for a model-predictive control.

FIG. 2 shows a first diagram.

The time after a failure (marked with an asterisk) is plotted in any units on the abscissa 100 of the first diagram. Negative values (intersection of abscissa 100 and ordinate 101) thus correspond to times before the failure.

The power of the respective component is plotted in any units on the ordinate 101 of the first diagram. Generation / provision is provided with a positive value and consumption / load with a negative value.

Before the failure, loads 204, 205 are covered by benefits 202, 203. The power 202 is provided by a CHP plant. The power 203 is provided by an electricity network. If the power grid now fails (power 203 drops to the value zero), critical load 205 must continue to be covered by contributing components of the energy system. In the present exemplary embodiment, the critical load 204 after the failure is covered by the CHP plant and a power 201 provided by an energy store. The non-critical load (output 205) is no longer supplied. The operating state of the contributing components is changed within the response time T _R , that is, the energy storage device and the CHP system are present. The energy storage and the CHP system cannot contribute the output ultimately required within the response time, but only part of their nominal output. According to the present invention, this is taken into account by the constraints that limit the change in performance over time from above. After the response time T _R and within the retrieval period r _D , the critical load is covered with constant power. This is symbolized by the straight line within the response time in the first diagram.

If the failure of the power grid 3 has been remedied, i.e. after the call time, a comparable adjustment of the operating points can take place, in which case the non-critical load - if still present - is covered again (not shown in the first diagram).

FIG. 3 shows a second diagram which illustrates an optimal solution, that is to say an optimal structure of an energy system, taking into account a failure function.

_{The call duration T O is} plotted in hours on the abscissa 100 of the diagram. The associated value of the sum of the objective function and the failure function is plotted on the ordinate 101.

The diagram is a bar diagram, with the striped bars corresponding to the value of the objective function and the unstriped bars corresponding to the value of the failure function. The sum of the two values thus corresponds to the sum of the bars mentioned.

The existence of an optimal structure with the lowest possible sum of target function and failure function can be understood based on two extremes of polling time.

If the call duration has the value zero, there are no additional components or redundancies to create a resilience of the energy system against a failure. The value of the objective function is therefore minimal in this regard. However, there is also a high risk of failure, since the energy system has no security in relation to failure. The default risk is quantified in this regard by the value of the default function. In other words, the value of the failure function is a maximum for this. As a result, there is typically no minimum of the total (sum of the two bars in the diagram) for this retrieval period.

For a call duration that tends towards infinity, that is, for the longest possible call duration, many additional components must ensure the resilience of the energy system over such a long period of time. For example, the energy system must have a plurality of energy stores, in particular special electrochemical stores. As a result, the risk of failure is minimal, but the value of the target function is maximal. This also does not correspond to a structure of the energy system that is as optimal as possible.

It is therefore necessary to find a compromise between the addition or the provision of components and the risk of failure. In other words, it is advantageous to determine the minimum of the sum of the target function and failure function as a function of the call duration. This minimum corresponds to the best possible structure with a certain retrieval time and the best possible efficiency. The mentioned minimum must be arranged between the extremes described above, and is marked with the arrow 242 in the present case. The method described can therefore be used to determine the best possible solution, that is to say the best possible structure of the energy system. Although the invention has been illustrated and described in more detail by the preferred exemplary embodiments, the invention is not restricted by the disclosed examples or other variations can be derived from them by the person skilled in the art without departing from the scope of protection of the invention.

List of reference symbols

1 energy system

2 critical system 3 power grid

20 Control unit 21 Energy store 22 Combined heat and power system 23 Emergency power system 41 Non-critical load

42 critical load 100 abscissa 101 ordinate 201 power energy storage 202 power CHP plant

203 Power grid

204 Performance critical load

205 Performance non-critical load

Claims

1. Computer-aided method for the numerical determination of the structure of an energy system (1) that is resilient to the failure of one or more of its components (21, ..., 23), the energy system (1) being described by an objective function intended for optimization , so that the structure of the energy system (1) is determined by calculating an extremum of the objective function, characterized in that the calculation of the extremum takes place under the constraint that a critical load P (42) is caused by one or more contributing components i ( 21, ..., 23) of the energy system (1) is _{further covered for a call duration T O in} such a way that components and / or processes dependent on the critical load P (42) are not significantly impaired, where the time change of the Power P _j ^{* of} each load-bearing component (21, ..., 23) within a response time T _{R of} the energy system (1) by the nominal power P _ne _{weighted with the start-up time T net i} t, £ of the respective component

(21, ..., 23) is restricted.

2. The method according to claim 1, characterized in that the temporal change in the powers P _j ^{* of} each contributing component (21, ..., 23) is limited _{by P n} et, i / T _n et _{, i.}

3. The method according to claim 1 or 2, characterized in that the power P _j ^{* of} each contributing component (21, ..., 23) within the response time T _{R of} the energy system (1) by P _j + Pnet _, iT _R / T _net i, where P _j denotes the performance of the respective component (21, ..., 23) before or during failure.

4. The method according to any one of the preceding claims, characterized in that the power P _j * of each contributing component (21, ..., 23) is _{limited within the response time T R} by their respective nominal power P _ne t, £.

5. The method according to any one of the preceding claims, characterized in that non-critical loads (41) are not covered during _{the retrieval period T O.}

6. The method according to any one of the preceding claims, characterized in that at least one of the contributing components (21, ..., 23) is an energy store (21), and that the calculation of the extremum takes place under the further secondary condition that the through the energy store (21) provided power P _ES during the call duration T _O satisfies the inequality? 7 _ES P _ES r _D <SOC Cap _ES , where h _{E $ is} the discharge efficiency of the energy store (21), SOC is the state of charge of the energy store ( 21) before or in the event of failure and Cap _ES denotes the capacity of the energy store (21).

7. The method according to any one of the preceding claims, characterized in that the calculation of the extremum under the secondary condition XiP _j ^* = P _E for all contributing components i (21, ..., 23) takes place _{during the retrieval period P D.}

8. The method according to any one of the preceding claims, characterized in that the critical load during the call duration from P _{D is set} by P _E = P (P _E ), where P _{E is} the performance of the critical load (42) before or in the event of failure denotes, and F describes the dependency of the components and / or processes dependent on _{the critical load P E (42).}

9. The method according to any one of the preceding claims, characterized in that the contributing components (21,

22) an energy storage device (21), a combined heat and power plant (22) and / or an emergency power system (23) can be used.

10. The method according to any one of the preceding claims, characterized in that the respective structure of the energy system (1) is determined _{for several retrieval periods P D.}

11. The method according to claim 10, characterized in that the respective value of the target function and a respective value of a failure function are determined for the plurality of specific structures, the failure function quantifying the failure risk of the critical load (42) as a _{function of the call duration T O, and} a structure is determined from the specific structures which has the smallest value of the sum formed from the target function and failure function.

12. Energy management system, comprising a control unit (20) for controlling a plurality of components (21, ..., 23) of an energy system (1), the control of the components (21, ..., 23) being provided for optimization Objective function based, so that by calculating an extremum of the objective function, control values for controlling the components

(21, ..., 23), characterized in that the control unit (20) is designed in such a way that the calculation of the extremum takes place under the secondary condition that a critical load (42) is caused by one or more contributing components i ( 21, ..., 23) of the energy system (1) for a retrieval period 7j _{) is} further covered in such a way that components and / or processes dependent on the critical load P (42) are not significantly impaired, with the change over time the power P _j ^{* of} each contributing component (21, ..., 23) within a response time T _{R of} the energy system (1) by the nominal power P _{net, £ of} the respective component (21, .. _{weighted with the start-up time T neti)} ., 23) is restricted.

13. Energy management system according to claim 12, characterized in that it has a data interface that enables data exchange between the energy management system and a local energy market, a cloud server and / or a decentralized data center.

14. Energy system (1) with several components (21, ..., 23), characterized in that the energy system (1) comprises an energy management system according to claim 12 or 13.

15. Energy system (1) according to claim 14, characterized in that the call duration T _{O of} the associated energy management system is set as a function of the location of the energy system (1).