Method and device for controlling the operation of an energy system

Abstract

The invention relates to a method for controlling the operation of an energy system, in which first target values x1 for the operation of the energy system are determined by means of a first optimization method by extremalizing a first target function z1. The method is characterized in that a second optimization method is carried out on the basis of an extremalization of a second target function z2 for determining second target values x2, the second target function z2 being determined using the difference between the maximum grid connection power PPCC,max (t) (4) of the energy system and an expected grid connection power PPCC (t;x2) (2) of the energy system. The secondary condition z1(x2) ≤ z1(x1) is used in the second optimization method, and the energy system is operated according to the determined second target values x2. The invention also relates to a device for controlling the operation of an energy system.

Description

[0001] 202411705 1 Description Method and device for controlling the operation of an energy system Modern energy systems can include on-site energy generation, for example, using photovoltaics or combined heat and power plants. Furthermore, these systems include storage systems, in particular battery storage or hot water storage. Electricity tariffs typically differentiate between energy-related components and peak load components. For electricity consumers, it is typically easier to reduce peak consumption during a billing period than to reduce total energy consumption. This is made possible in particular by the flexibility of the energy system. Storage facilities, generation plants, and loads can be operated flexibly, allowing peak loads to be shifted to off-peak times. This can reduce price peaks.Furthermore, this can relieve the strain on the grid infrastructure, and grid expansion measures, such as increasing transformer or line capacities, can be reduced or even avoided. Optimally controlling energy systems is a complex technical challenge that is particularly difficult to overcome with heuristic rules. Instead, model-based energy management systems (EMS) offer a promising approach to optimally controlling complex energy systems. Such a model-based EMS can retrieve tariff forecasts and generate upcoming load and generation forecasts, which are used together with the EMS's energy system model to determine the most optimal setpoints for the energy system's components. The determination of these setpoints is achieved using an optimization procedure based on an objective function. The objective function depends on the setpoints.As part of the optimization process, the target values ​​are determined in such a way that the objective function is extremized as much as possible. In this sense, the target values ​​determined in this way are as optimal as possible. The objective function, together with the boundary conditions, forms an optimization model of the energy system, which typically operates the energy system close to its technical limits, for example, close to a maximum permissible grid connection capacity or close to the rated capacities of plants or equipment. This typically leads to high efficiency. However, this also entails a greater risk of unforeseen load peaks. This is particularly the case if load and / or generation forecasts are too inaccurate. The present invention is based on the objective of providing improved operation for an energy system, in particular to reduce unforeseen load peaks and / or to respond adequately to them.The problem is solved by a method with the features of independent claim 1 and by a device with the features of independent claim 11. Advantageous embodiments and further developments of the invention are specified in the dependent claims. The method according to the invention for operating an energy system, in which first setpoints for the operation of the energy system are determined by means of a first optimization method by extremalizing a first objective function. The method, which is determined, is characterized in that a second optimization procedure based on extremalizing a second objective function ^^2 is carried out to determine second setpoints ^^2, wherein the second objective function ^^2 is formed by means of a difference between a maximum grid connection power ^^PCC ,max (^^) of the energy system and an expected grid connection power ^^PCC (^^;^^2) of the energy system, wherein the constraint ^^1(^^2) ≤^^1(^^1) is used in the second optimization, and the energy system is operated according to the determined second setpoints ^^2. The method according to the invention and / or one or more functions, features and / or steps of the method according to the invention and / or one of its embodiments can be computer-aided. In particular, the optimizations or the optimization procedures can be carried out numerically using a computing unit.The term "control" in this context also includes "regulation." In particular, the method according to the invention is a component of a model predictive control system and / or an energy management system of the energy system. The setpoints are, in particular, the outputs of energy systems within a time range, for example, within one hour or within 15 minutes. In other words, a method according to the invention is carried out for several such time ranges, and setpoints for controlling the energy system systems are determined for each of the time ranges, for example, in fifteen-minute time steps. Alternatively or additionally, for a longer time range, for example, a day, the method can be carried out with a temporal resolution of one hour and / or 15 minutes. The respective setpoints are summarized here in the form of a setpoint vector ^^1 or ^^2.The first objective function is fundamentally a function in terms of variables, or rather, in terms of general setpoints, particularly power outputs. In other words, 1 = 1(1). Thus, 1(1 = 1) = 1(1), meaning 1(1) is the optimal value of the first objective function according to the first optimization; that is, the value the first objective function has when the determined initial setpoints are substituted. From a structural perspective, the IPCC Fifth Assessment Report, in particular, defines an energy system as: "All components relating to the generation, conversion, supply, and use of energy" (Annex I, page 1261). Energy systems typically comprise several components, especially energy infrastructure facilities, such as energy conversion plants, consumption facilities, and / or storage facilities. Multimodal energy systems are energy systems that generate and / or supply multiple forms of energy.In particular, a multimodal energy system provides one or more forms of energy to an energy consumer, such as a building, an industrial plant, or private facilities. This provision is achieved primarily through the conversion of different energy forms, the transport of different energy forms, and / or the storage of energy forms. In other words, the various forms of energy, such as heat, cooling, or electrical energy, are coupled by means of the multimodal energy system with regard to their generation, provision, and / or storage.As installations or energy systems, the energy system can comprise one or more of the following components: power generators, combined heat and power plants, in particular cogeneration units, gas boilers, diesel generators, heat pumps, compression chillers, absorption chillers, pumps, district heating networks, energy transmission lines, wind turbines or wind power plants, photovoltaic systems, electricity storage systems, biomass plants, biogas plants, waste incineration plants, industrial plants, conventional power plants, and / or the like. Optimization within the meaning of the invention is a method for extremization, that is, for minimizing or maximizing an objective function. Minimizing or maximizing the objective function is typically extremely complex and is therefore usually performed numerically. The objective function models a property or quantity of the energy system, for example, its total carbon dioxide emissions and / or its operating costs.The objective function has parameters and variables. The result of the optimization is the values ​​of the variables, in this case, the target values ​​for the operation of the energy system or its components. The parameters are fixed and parameterize the objective function specific to the energy system. Typically, the optimization process does not aim to achieve an exact minimum or maximum of the objective function; rather, it is sufficient to approximate it sufficiently, for example, by setting a threshold value. Furthermore, the optimization is typically carried out taking several constraints into account. Constraints, boundary conditions, or constraints—collectively referred to here as constraints—are conditions, properties, and / or relations that the parameters and / or variables of the optimization process must satisfy.These can be given as equations and / or inequalities, and / or explicitly describe a set of permissible values ​​of the parameters and / or permissible values ​​of the variables. According to the inventive method, two optimization procedures are carried out. In other words, a multi-stage optimization procedure is performed. The second optimization can be referred to as lexicographic optimization. Using the first optimization procedure, the first target values ​​are determined. For this purpose, the first objective function or its value is extremalized, that is, minimized or maximized. Within the framework of the second optimization, or using the second optimization procedure, the second target values ​​are determined.The second setpoints are intended for the operation of the energy system, or for controlling its operation, at least within a specific timeframe, for example, within the next 15 minutes. The second optimization is based on a second objective function, which is defined by the difference between the maximum grid connection capacity of the energy system and the expected grid connection capacity. The maximum grid connection capacity is the maximum permissible grid connection capacity of the energy system at at least one grid connection point (Point of Common Coupling; PCC). The expected grid connection capacity is the total power or load that is expected to result from the operation of the energy system, particularly its components. Therefore, the expected grid connection capacity depends on the second setpoints.202411705 5 According to the second optimization, the difference between the maximum grid connection capacity of the energy system and the expected grid connection capacity of the energy system is extremalized, that is, maximized without restricting the generality and / or the scope of protection of the invention. This ensures that the operation of the energy system, or the total output of the energy system at the grid connection point, maintains the greatest possible, or optimal, distance from the permissible grid connection capacity. As a result, the energy system is typically not operated at its operating limits. Advantageously, this creates a safety margin or buffer relative to the maximum permissible grid connection capacity.However, the operation of the energy system, or rather its control, remains optimal with respect to the first objective function or the first optimization. This is ensured by the fact that, according to the invention, the second optimization is performed under the constraint that the value of the first objective function in the second optimization is less than or equal to the value of the first objective function determined by the first optimization method. In other words, the constraint μ1 ≤ μ1(μ1) = μ̃^1 is used in the second optimization. This means that the first objective function, which is formed in the second optimization using the second setpoints as variables, is less than or equal to the value of the first objective function determined in the first optimization according to the first setpoints. This applies if the first objective function is minimized without limiting the generality and / or scope of the invention.To maximize the first objective function, the following would result analogously: ^^1 ≥ ^^1(^^1) = ^̃^1. Subsequently, without limiting the generality and / or scope of the invention, the first objective function is minimized and the second objective function is maximized. The value ^̃^ is derived from the first objective function by substituting the initial setpoints into it. This substitution of the initial setpoints into the first objective function is symbolized by ^^1(^^ = ^^1) = ^^1(^^1) = ^̃^1. In other words, the constraint ^^1(^^ = ^^2) = ^^1(^^2) ≤ ^̃^1 is used. Therefore, the value of the first objective function cannot be further worsened during the second optimization.The second optimization therefore seeks a solution, or rather, second setpoints, that are at least as optimal as the first setpoints with respect to the first objective function, but which, in addition to the grid connection capacity, exhibit a safety buffer of the largest possible size with respect to their total power, i.e., a difference of the largest possible size from the maximum permissible grid connection capacity. The control method according to the invention thus operates the energy system as optimally as possible (first objective function) and simultaneously reduces its power, or rather its total power, at its grid connection point compared to the maximum grid connection capacity. This advantageously creates a safety buffer, so that power peaks can also be reduced. In other words, load peaks, which place additional strain on the grid infrastructure, are reduced and / or avoided.The device according to the invention for controlling the operation of an energy system comprises a control unit and a computing unit, wherein the control unit is configured to control the operation of the energy system according to setpoints, and the computing unit is configured to determine first setpoints ^^1 for the operation of the energy system by means of a first optimization method by extremizing a first objective function. to determine. The device according to the invention is characterized in that the computing unit is configured to perform a second optimization method based on extremizing a second objective function ^^2 to determine second setpoints ^^2, wherein the second objective function ^^2 is formed by means of a difference between a maximum grid connection power ^^PCC ,max (^^) of the energy system and an expected grid connection power ^^PCC (^^;^^2) of the energy system, wherein the constraint ^^1(^^2) ≤^^1(^^1) is used in the second optimization, and the control unit is configured to control the operation of the energy system according to the determined second setpoints ^^2. The device according to the invention can, in particular, be configured as an energy management system of the energy system. Similar, equivalent, and equivalently effective advantages and / or embodiments of the device according to the invention are obtained in relation to the method according to the invention.According to an advantageous embodiment of the invention, the second objective function ^^2 = ^^2(^^2) is formed by means of ^^2 = ∑^^ [^^PCC ,max (^^) − ^^PCC (^^;^^2)], wherein the second objective function ^^2 is maximized within the framework of the second optimization procedure. Advantageously, the distance to the maximum permissible grid connection power ^^ is taken into account. PCC ,max ( ^^ ) maximized. In other words, the safety buffer (difference) for the operation of the energy system is maximized. Analogously, −^^2 can be minimized. 202411705 7 The second optimization thus attempts to maximize the safety buffer as the difference between the maximum power at the grid connection point ^^PCC ,max (^^) and the expected power^^ PCC ( ^^;^^2 )to maximize at the network connection point while preserving the objective function defined in the original optimization. Furthermore, the lexicographical optimization step, i.e., the second optimization, can also include terms for minimizing additional, secondary objectives, such as gradient minimization. In an advantageous embodiment of the invention, the first objective function is determined by means of = ^^^^ ⋅ ^^1, where ^^ is a parameter vector. Advantageously, this provides a linear objective function. This allows the first optimization to be performed more efficiently. Here, ^^^^ ⋅ ^^1 is the scalar product between the parameter vector ^^ and the vector of variables / setpoints ^^ or the first setpoints ^^1. In other words, the first optimization problem can be symbolized by minimize ^^1 =^^^^ ⋅ ^^, such that ^^1 = argmin[^^1(^^)]. According to an advantageous embodiment of the invention, further constraints of the form ^^ ⋅ ^^1,2 ≤ ^^ are considered in the first and / or second optimization method. Advantageously, this allows further technical boundary conditions, in particular technical boundary conditions of the plants, for example, maximum rated power, to be considered as constraints in the optimizations. Here, ^^ is a matrix and ^^ is a vector.In particular, the first optimization can thus be written as minimize ^^ ^^1 = ^^ ⋅ ^^, subject to ^^ ⋅ ^^ ≤ ^^. The second optimization can be written as maximize ^^2 = ∑^^ [^^PCC ,max (^^) −^^ (^^;^^)], subject to ^^ ^^PCC ect to ^^ ⋅ ^^ ≤ ^^ and subject to ^^ ⋅ ^^ ≤ ^̃^1 = ^^ ⋅ ^^1, where ^^2 = argmax[^^2(^^)]. In an advantageous embodiment of the invention, the first objective function ^^1 models the total carbon dioxide emissions of the energy system. Advantageously, this allows the total carbon dioxide emissions of the energy system to be reduced during its operation. In this case, the parameter vector ^^ is specifically determined by carbon dioxide emissions, while the variable vector ^^ and thus also the setpoints ^^^^ and ^^^^ are determined by the performance of the plants.In other words, ^^1 = ∑ ^^ ^^^^;^^ ^^ ^^^^;^^ where ^^ is the specific carbon dioxide emission, for example kilograms per kilowatt hour, of the ^^-th plant of the energy system and ^^^^;^^ is the time-dependent power of the ^^-th plant of the energy system. According to an advantageous embodiment of the invention, the maximum grid connection power ^^PCC ,max (^^) is reduced for the second optimization method in time periods in which power above a first threshold value is expected. This advantageously reduces power peaks. Thus, a new (synthetic / virtual), smaller maximum grid connection power is used for the second optimization method, which is reduced compared to the actual maximum grid connection power. This reduces peak power and / or shifts it to non-critical time periods.Here, the first threshold can correspond to or be less than the maximum grid connection capacity. This corresponds to a dynamic buffer allocation that provides additional time-variable operational reserves (buffers) to handle unexpected load peaks. This reduces operating costs and the strain on the grid infrastructure. Furthermore, grid expansion measures, such as the modernization of lines and / or transformers, are reduced or can even be completely avoided. In an advantageous embodiment of the invention, the expected power is determined by means of a load forecast (PCC). This advantageously improves the aforementioned avoidance or reduction of power peaks. Statistical forecasts and / or historical data can be used for the load forecast. The forecast preferably has a temporal resolution of 15 minutes.According to an advantageous embodiment of the invention, the maximum grid connection power ^^PCC ,max (^^) is reduced in a time range in which a power peak ^^PCC ,peak above the first threshold value according to ^^PCC ,forcast (^^) is expected. formed, where ^^ is a safety margin. This advantageously further improves the aforementioned avoidance or reduction of power peaks. 202411705 9 In an advantageous embodiment of the invention, the maximum grid connection power ^^PCC ,max (^^) is increased for the second optimization method in time ranges in which power below a second threshold is expected. This particularly improves the shifting of power peaks to non-critical time ranges. In other words, ^^PCC ,max (^^) of the second optimization problem is (synthetically) adjusted such that it is higher at times outside of peak times and lower during the expected peak times. Further advantages, features, and details of the invention will become apparent from the exemplary embodiments described below and from the drawings.Figure 1 schematically shows a flowchart of a method according to an embodiment of the invention; and Figure 2 shows a diagram relating to the total output of an energy system. Similar, equivalent, or equivalently functioning elements may be designated with the same reference numerals in one or more of the figures. Figure 1 shows a flowchart of a method for controlling an energy system according to an embodiment of the invention. Here, the control of the energy system, in particular of a building, is based on the control of its energy systems, whereby setpoint values ​​for the respective output of the systems within a time range are specified. These setpoint values ​​are determined by the method using a first and second optimization, or by a first and second optimization process. According to a first step S1 of the method, first setpoint values ​​^^1 are determined by a first optimization.Here, the first optimization is based on a first objective function ^^1. The first objective function can be in the form. = ^^1(^^) = ^^^^ ⋅ ^^, where ^^ is a parameter vector that models the plant-technical structure of the energy system. Here, ^^ are the variables of the objective function, that is, in this case, the target powers or the setpoints. The first setpoints ^^1 thus form a solution of the first optimization, meaning that ^^1 = argmin[^^1(^^)] holds true if the first objective function is minimized within the framework of the first optimization and thus for determining the first setpoints ^^1. 202411705 10 In a second step S2 of the procedure, the second optimization takes place. The second optimization is based on a second objective function ^^2 = ^^2(^^) = ∑^^ [^^PCC ,max (^^) − ^^PCC (^^;^^)] , which is maximized within the framework of the second optimization. Here, ^^PCC ,max (^^) is the maximum permissible grid connection power of the energy system and ^^PCC (^^;^^) is the total power expected at the grid connection point due to the operation of the energy system.The result, or rather the solution, of the second optimization is the second setpoint value ^^2. In other words, ^^2 = argmax[^^2(^^)]. The second optimization also includes the constraint that the value of the first objective function determined by the first optimization must not worsen, meaning it must not be increased. In other words, the constraint of the second optimization is ^^1(^^) = ^^^^ ⋅ ^^ ≤ ^^1(^^1) = ^̃^1 with ^̃^1 = ^^1(^^ = ^^1). The second optimization thus metaphorically searches the solution space for a solution that maximizes the difference between the total power output and the maximum grid connection capacity, thereby providing the largest possible safety margin, while leaving the operationally optimal solution from the first optimization unchanged or improving upon it. Therefore, the second optimization is a lexicographical optimization.In particular, as part of the second optimization to avoid peak loads, the maximum grid connection capacity can be synthetically adjusted, meaning it can be reduced or increased compared to the actual maximum permissible grid connection capacity. This corresponds to dynamic buffer allocation. In other words, dynamic buffer allocation can also be performed to avoid power peaks. This requires a prediction of the power peaks, for example, by means of simulated control of the energy system. The setpoints are not optimized in this process. For an energy system that includes, for example, a photovoltaic system (PV system), a battery storage system, and electrical loads (consumers), this could, for instance, be a control strategy without using the battery storage system.The power profile at the grid connection point would therefore be the difference between electrical load and PV generation. At the grid level, this could lead to a well-known "duck curve." The duck curve is characterized by a slightly increased load in the morning, a sharp drop around midday and early afternoon, and a distinct peak in the early evening. The simulation described above results in such power curves ^^PCC ,forcast. ( ^^ )for the local grid connection point of the energy system. The simulation thus allows an estimation of the timing and amplitude of expected power peaks. This information can be used for the present control system. Here, ^^PCC ,max (^^) is adjusted for the second optimization problem such that it is higher at times outside of power peaks and lower during the expected power peaks. A preferred adjustment of ^^PCC ,max (^^) is given by the following equation, where ^^PCC ,peak represents the previously generated peak consumption in the current billing period and ^^ represents a safety margin to avoid an increase in the previous peak consumption: According to a third step S3 of the procedure, the second setpoints ^^2 are ultimately used for the operation of the energy system's components, i.e., for controlling the operation of the components. These second setpoints are typically time-dependent, meaning they can have different values ​​for different time periods. For example, the first and / or second setpoints are defined as a time series with a temporal resolution of one hour, particularly 15 minutes. Figure 2 shows a diagram of the total power output of an energy system. Time in arbitrary units is plotted on the abscissa 100 of the diagram. The (total) power output of the energy system at its grid connection point is plotted in kilowatt-hours (kW) on the ordinate 101 of the diagram. The power output, or power curve, or total power output of the energy system is indicated in the diagram by the reference symbol 1.A forecast of the energy system's power output at its grid connection point is indicated by the dashed line 2. A maximum permissible grid connection power of the energy system at its grid connection point is indicated by the line 4. In this case, the maximum grid connection power thus has an exemplary value of 30 kW. Within a critical time range 3, the energy system, according to forecast 2, exhibits a peak power output above the maximum permissible grid connection power 4. The present invention, or one of its embodiments, makes it possible, through the second optimization, to reduce these peak loads, so that, compared to forecast 2, the actual power profile 1 shows a smaller exceedance of the maximum permissible connection power.This allows the operating costs of the energy system to be reduced and the electricity grid, to which the energy system is connected via its grid connection point, to be relieved of some of its load. Although the invention has been illustrated and described in detail by the preferred embodiments, the invention is not limited by the disclosed examples, nor can other variations be derived from them by a person skilled in the art without departing from the scope of protection of the invention.

[0002] 202411705 13 Reference Symbol List 1 Grid Connection Capacity (Total Capacity) 2 Load Forecast 3 Critical Time Range 4 Maximum Grid Connection Capacity 100 Abscissa 101 Ordinate S1 First Step S2 Second Step S3 Third Step

Claims

202411705 14 Claims 1. Method for controlling the operation of an energy system, wherein first setpoints ^^1 for the operation of the energy system are determined by means of a first optimization method by extremizing a first objective function 1. The method according to claim 1, characterized in that a second optimization procedure based on extremalizing a second objective function is carried out to determine second setpoints, wherein the second objective function is formed by means of a difference between a maximum grid connection power PCCmax(4) of the energy system and an expected grid connection power PCC2(2) of the energy system, wherein the constraint 1(2) ≤ 1(1) is used in the second optimization, and the energy system is operated according to the determined second setpoints.

2. The method according to claim 1, characterized in that the second objective function is formed by means of 2 = ∑ [PCCmax(2) − PCC2(2)], wherein the second objective function is maximized within the framework of the second optimization procedure. 3.A method according to claim 1 or 2, characterized in that the first objective function is formed by means of ^^1 = ^^^^ ⋅ ^^1, where ^^ is a parameter vector.

4. A method according to any of the preceding claims, characterized in that, in the first and / or second optimization method, further constraints of the form ^^ ⋅^^1,2 ≤ ^^ are taken into account.

5. A method according to any of the preceding claims, characterized in that the first objective function ^^1 models the total carbon dioxide emissions of the energy system.

6. A method according to any of the preceding claims, characterized in that the maximum grid connection power ^^PCC ,max (^^) (4) is reduced for the second optimization method in time ranges (3) in which power above a first threshold value (4) is expected.

7. A method according to claim 6, characterized in that the expected power is determined by means of a load forecast ^^PCC ,forcast (^^) (2). 8.Method according to claim 7, characterized in that the maximum grid connection power ^^PCC ,max (^^) (4) in a time range (3) in which a power peak ^^PCC ,peak above the first threshold value (4) according to ^^PCC ,forcast (^^) (2) is expected, is by. 202411705 15 is formed, where ^^ is a safety distance.

9. Method according to one of the preceding claims, characterized in that the maximum grid connection power ^^PCC ,max (^^) (4) is increased for the second optimization method in time ranges in which power below a second threshold value is expected.

10. Device for controlling the operation of an energy system, comprising a control unit and a computing unit, wherein the control unit is configured to control the operation of the energy system according to setpoints, and the computing unit is configured to determine first setpoints ^^1 for the operation of the energy system by means of a first optimization method by extremizing a first objective function. zucharacterized in that the computing unit is configured to perform a second optimization procedure based on extremalizing a second objective function ^^2 to determine second setpoints ^^2, wherein the second objective function ^^2 is formed by means of a difference between a maximum grid connection power ^^PCC ,max (^^) (4) of the energy system and an expected grid connection power ^^PCC (^^;^^2) (2) of the energy system, wherein the constraint ^^1(^^2) ≤ ^^1(^^1) is used in the second optimization, and the control unit is configured to control the operation of the energy system according to the determined second setpoints ^^2.

Images