EP2697753A1 - Adjustment of an industrial installation - Google Patents

Abstract

It is proposed that an industrial installation, e.g. a production or logistics system, be operated in optimized fashion, with the power consumption being optimized for a target function. Besides the power consumption, the target function may also factor in further parameters, e.g. for an energy supplier or the installation itself, with the result that multitarget optimization, for example, can be performed and the installation can be adapted in respect of the power consumption or the power consumption can be adapted in respect of the installation. Both the industrial installation and an energy supplier or network operator provide information which can be factored in as appropriate for the purpose of optimization or as part of the target function. In this case, it is advantageous that overload situations are avoided and, in particular, a large number of regenerative energy sources can be used as energy suppliers, because adaptation is effected in line with the amount of energy actually provided and hence the power supply system can be operated and loaded as appropriate. The invention can be used in smart grids or in production or logistics management systems, for example.

Description

description

Setting up an industrial plant The invention relates to a method and a device for setting an industrial plant. In particular, the invention allows an efficient use of smart grids or advantageous utilization of production or lodging ^¬ policy management systems.

EPCIS (English for Electronic Product Code Information ^Services ) is a standard that was published in its first version in 2007 and essentially defines interfaces for the collection and retrieval of so-called EPCIS events. EPCIS empowers users (businesses, government, supply chains, etc.) to increase visibility and control over their respective processes. EPCIS can both within the company and across companies is ^¬ sets (see: http://de.wikipedia.org/wiki/EPCIS).

The term intelligent power grid (also referred to here as a smart grid) comprises the communicative networking and control ^¬ tion of power generators, storage, electrical loads and network resources in energy transmission and - distribution networks of the electricity supply. This type of networking enables optimization and monitoring of the connected components. The aim here is to ensure the energy supply based on efficient and reliable system operation (see:

http://en.wikipedia.org/wiki/Smart_Grid).

The increasing expansion of renewable energies and the resulting increase in fluctuating, uncontrollable and therefore unpredictable power feeds into a power supply network (also referred to as power grid) endangers the

Security of supply and grid stability. Frequently one ent speaking ^¬ expansion of the power grid is neither feasible nor advisable timely and economically. For this reason, alternative strategies for efficient use of the power grid are increasingly proposed. One way to make network utilization more efficient is to introduce additional information technology (IT) into the power grid; such a power grid is often referred to as an "intelligent power grid" or as a "smart grid". Based on such an IT infrastructure, coordination mechanisms can be used to increase network utilization and / or network stability. However, the development of suitable coordination mechanisms is in many cases not yet available and the corresponding concepts are usually expensive to implement as a new infrastructure for coordinating and controlling consumers is needed.

Another disadvantage is that the known coordination mechanisms are used for devices that are easy to control, such as air conditioning systems or heating pumps. Coordination of large consumers such as industrial plants is much more complex and requires different and possibly more elaborate coordination mechanisms ^¬.

The object of the invention is to avoid the above disadvantages and in particular ge ^¬ called to create a possible ^¬ ness for the efficient control of an industrial or technical plant, for example by optimizing the power consumption or the cost of power consumption.

This object is achieved according to the features of the independent claims. Preferred embodiments are in particular the dependent claims.

To solve the problem, a method for setting an industrial plant is proposed, - in which a power consumption of the industrial plant is determined,

- in which an optimization of power consumption is carried out in ^¬ industrial plant with respect to a target function,

- in which the industrial system is set according to the Op ^¬ optimization.

For example, the power consumption of the industrial plants can be adjusted according to the optimization. It is also possible that / industrial complex itself, eg ei ^¬ ne processing speed and or processing modes are set according to the optimization. Thus, the solution proposed here beispielswei ^¬ se enables energy-optimized production planning and control for industrial plants, such as production equipment, logistical systems, etc. based on existing logistics and production logistics systems.

Advantageously, the power supply can be considered analogous to just-in-time or just-in-sequence deliveries of vendor parts to manufacturing companies and the production control systems for the realization of supply-demand effects in a smart grid can be extended accordingly.

This means for example that a single system for the procurement of materials (eg raw / auxiliary material) or goods for production and for the procurement of the need for the production processes of energy used who can ^¬.

In this case, it is advantageous, for example, that due to the existing logistical or production logistics systems, no additional information systems are needed with regard to the energy consumer (manufacturing company) and thus are less complex. costs for the introduction of smart grid solutions.

Furthermore, it is advantageous that the energy costs of the energy consumer by flexible adaptation of the system, eg production, for example by energy-optimized order ^¬ control or machine configuration, to external events (eg price signals) can be reduced. It is also an advantage that energy suppliers (eg energy network operators) receive additional flexibility because events make it possible to control the requested energy. This avoids inefficient grid expansion, which is required only in rare cases. Furthermore, the share of fluctuating energy suppliers (eg regenerative energy suppliers) can be increased.

A development is that the objective function takes into account information regarding past power consumption.

In particular, historical data can be collected and taken into account as part of the optimization. This has the advantage that, for example, a prognosis can be carried out based on the historical data. So it is possible to predict an outcome, such as a delay in completion of a product or a corresponding change in the power consumption of industriel ^¬ len facility and appropriate and timely countermeasures ^¬ took to take. Another development is that the objective function takes into account information from an energy supplier, in particular a signal relating to the power consumption.

In particular, it is a further development that the information from the energy supplier comprises at least one of the following information:

a signal relating to the minimum and / or maximum power consumption to be set; - a price information;

 - another fare information;

 - a utilization information. It is also a development that the objective function information of the power grid, in particular a state of the power grid, taken into account.

This information concerning the power grid can be determined by the industrial plant or another measuring unit.

Furthermore, it is a development that the objective function takes into account a current, a past and / or a planned power consumption of the industrial plant.

In the context of an additional development, the optimization of power consumption and at least one further Para ^¬ meters of the industrial installation is performed with respect to the objective function.

A next development consists in that the at least one further parameter comprises:

 - a parameter of the industrial plant;

 - a lead time;

 - costs of the system;

 - cost of electricity;

 - a logistics plan;

 - a production plan.

One embodiment is that the objective function takes into account a aktuel ^¬ le or future situation of the industrial plant. The future situation may be a predicted situation of the industrial plant. An alternative embodiment is that the ak ^¬ tual or the future situation of the industrial plant at least one of the following information into account:

 - a delivery status of delivered or provided goods;

 - an expansion status of the industrial plant;

 - an operating mode of the industrial plant.

A next embodiment is that the objective function comprises at least one of the following objectives:

 - compliance with a delivery commitment;

 - compliance with a given machine utilization;

- compliance with a given energy consumption;

 - compliance with a throughput.

The objective function may be a cost function, in particular a weighted cost function. The objective function can be optimized for a goal or goals in terms of several (Einziel- or multi-objective optimization Opti ^¬).

It is also an embodiment that the industrial plant comprises at least one of the following components:

 - a production or manufacturing;

 - a logistics unit;

 - a service unit;

 - a data center;

 - an energy supplier;

 - a transmission system operator.

In particular, the industrial plant may be any plant which requires a significant amount of electricity for the energy supplier. The industrial plant may e.g. include one or more companies.

The above object is also achieved by a device for setting an industrial plant comprising Send a processing unit that is set up so that

 a power consumption of the industrial plant can be determined,

- optimizing the power consumption of industriel ^¬ len plant is feasible in terms of an objective function,

 - The industrial plant is adjustable according to the optimization.

The processing unit mentioned herein that may be embodied in particular as a processor unit and / or an at least partially solid ^¬ wired or logic circuitry, for example, is set up such that the method can be performed as described herein. said

Processing unit may be any type of processor or computer or computers with corresponding peripheral systems (SpeI ^¬ cher, input / output interfaces, input-output devices, etc.) or include.

The above explanations regarding the method apply to the device accordingly. The device may be implemented in one component or distributed in several components.

A development is that the device is part of a production management system.

The object mentioned above is also achieved by means of a system comprising at least one of the ^devices described here.

The presented solution further comprises a Computerpro ^¬ program product, directly loadable into a memory of a digital computer, comprising program code portions which are suitable to carry out steps of the method described herein. Furthermore, the above problem is solved by means of a computer-readable storage medium, eg of any memory, comprising computer-executable instructions (eg in the form of program code) suitable for the computer to perform steps of the method described herein.

The above-described characteristics, features, and advantages of this invention, as well as the manner in which they will be achieved, will become clearer and more clearly understood in connection with the following schematic description of exemplary embodiments which will be described in detail in conjunction with the drawings. For the sake of clarity, identical or equivalent elements may be provided with the same reference numerals.

Show it:

Fig.l a manufacturing company as a customer of

 Energy from an energy supplier;

2 shows an event processor as part of the assistance system shown in FIG. 1, wherein the event processor comprises a registration, a situation recognition and a compensation;

FIG. 3 is an algorithm describing how to log requests to an EPCIS database event mechanism and how to provide the required availabilities of the situation detection resources; FIG.

4 is a schematic diagram illustrating the dependencies between production jobs, removal jobs, and delivery jobs.

The approach proposed here makes it possible an energy consumption ^¬ a complex system, such as a manufacturing Company or a service provider to influence so that utilization and / or stability of the Energieet ^¬ zes can be improved. This is advantageously achieved by Warenlogistik- and supply chain management solutions for a "just right ^¬ term" (Just-In-Time (JIT) or Just-In-Sequence (JIS)) An ^¬ delivery, processing and / or control is used accordingly.

Here, use ^¬ to that flow for JIT / JIS delivery strategies real-time events in the production logistics systems beneficial effect or trend. Thus, existing systems (eg production logistics

Systems) often respond to external events by the planning and / or controlling the production and Logistikpro ^¬ processes will be adjusted / be. Such systems are, for example, a

 - Supply Chain Management (SCM),

 - Enterprise Resource Planning (ERP) and or

 - Manufacturing Execution System (MES).

For an explanation of the above terms, see [http://de.wikipedia.org/wiki/Supply-Chain-Management], [http: // de. wikipedia. org / wiki / Enterprise_Resource_Planning] and

 [http: // de. wikipedia. org / wiki / Manufacturing_Execution_System].

For example, a smart grid can provide dynamic price signals. Price changes can be used as events to achieve energy-optimized planning or energy-optimized control of electricity consumption.

In particular, because of their typically high energy consumption (and their thus significant impact on the electricity grid) is an important factor in the introduction of the Smart Grid concept.

By using existing IT infrastructures of the JIT / JIS logistics control, a relatively cost-effective solution for the implementation of a so-called "demand-response" scenario in or for the smart grid can be realized.

Fig.l shows a manufacturing company 101 as a buyer of a considerable amount of energy from a power supplier 102nd

Both the company 101 and the energy supplier 102 provide an event server 103 with information 108, 109, which are processed as control events 104 by an assistance system 105. The assistance system 105 includes at least one of the aforementioned systems in ^¬ way of example: SCM, ERP, MES. The enterprise 101 provides the event server 103 as information 108 with, for example, delivery information, status information regarding delivery or the like. The energy supplier ^¬ 102 sets the event server 103 as Informatio ^¬ NEN 109, for example, price information, supply-demand information or energy consumption plans or amendments thereto are available.

For example, in the case of a JIT delivery after an order that is already carried out early, a delivery or special call-off takes place (eg according to VDA 4905/4915/4916), which will be discontinued at short notice as required and determines the required quantity and the time. The corresponding IT systems (ERP / SCM / MES) in the assistance system 105 can respond flexibly to other delivery quantities or other delivery times (eg by rescheduling). Accordingly, the Assis ^¬ assistance system 105 the company 101 106 feedback concerning eg order and / or delivery schedule ready. If it is in the company 101, for example, a major customer of the energy supplier 102, a long-term supply (according to the ordering of goods) best ^¬ hen, the short-term roadmap predictions can concerning the loading urged electricity a day destined (according to a delivery schedule for goods) compare information 107 provided by the assistance system 105 to the energy supplier 102. Preferably, a smart grid can have the function that, depending on the state of the power grid and the power generation ^¬ forecasts the possible consumption both a flexible electricity price (event notification) and a corresponding supply-demand signal, which is regulated for example in a contract, is controlled.

Further advantages and implementations:

For example, it allows the approach presented here, existing logistical and / or production logistics syste ^¬ me in particular producing industries inexpensively integrated into a smart grid. For example, (dynamic) price signals and supply-demand strategies can be used to efficiently source electricity from the energy supplier.

Here are two possible scenarios he ^¬ explained by way of example: (1) For example, logistical events can be replaced, for example, in ei ^¬ nem companies and / or business-to etc an event-based exchange of energy information (eg electricity price changes, energy demand changes, schedule changes, .).

To this end, over ^¬ monitoring system (also referred to as tracking and ordered an existing (eg RFID-based) Cing system) are used, for example, for supply chains may even already exist and can be used in the context of the present invention.

Preferably, an information model is augmented by events (e.g., by extension of the EPCIS standard) needed to control power delivery. These may be, for example, price signals or offer-and-demand signals (see information 109).

(2) An existing (eg, production logistics) Assisting ^¬ assistance system 105 can be extended in such a way that it, for example, current provisioning events (for example, transmitted in the form of information 109) receives and, if required, the production planning (and hence the power consumption) influences (see FIG. Feedback 106 to company 101).

The approach thus makes it possible to respond to logistic events (information 108) as well as energy information 109 and to optimize power consumption based on energy models of the affected industrial plant (i.e., company 101)

 a rescheduling of the production (for example a production order rescheduling),

- a change in the production processes (eg Variegated ^¬ tion of the speed of production) and / or

- a dynamic change of the electricity supplier and / or electricity tariff.

In particular, the approach presented here can track at least ei ^¬ nes the following objectives:

(a) Total energy consumption or consumption of non-regenerative energy can be minimized. For example, jobs can be relocated in times when a large amount of wind or solar energy is available. (b) Peak loads can be reduced or avoided. This has the particular advantage that a smaller Capa ^¬ capacity of the electric wires is sufficient.

(c) It can flexibly respond to an actually existing and be ^{¬ ¬} reitstellbare amount of electrical energy to. In particular, excess or reduced amounts of electrical energy can also be at least partially compensated with regard to an energy timetable.

(d) A timely change between electricity suppliers and / or electricity tariffs can increase cost efficiency.

For example, an RFID based production logistics system can be optimized as described herein. RFID enables automatic identification and Lo ^¬ delocalization of objects and thus facilitates Erfas ^¬ solution of data (see. Http://de.wikipedia.org/wiki/RFID).

One example is the research project RAN

(www.autoran.de), which uses a logistic system with RFID-based monitoring (tracking and tracing system) of objects and the corresponding production-logical assistance systems for event-based production planning and control.

Any production logistics infrastructure can be used for energy-optimal planning and control of a manufacturing operation. Here, events of the electricity supplier (e.g., price change), distribution network operator (e.g., congestion notification), etc., may be appropriately responded.

The energy suppliers (energy), such as network operators ^¬ over, electricity distributors, transmission system operators, distribution system operators, meter operators may have different energy-related events, such as a congestion event, a rate change event, a consumption status event, etc. provide the event server, which then transmits a control event to the assistance system.

On the company side, different energy consumers can participate, who transmit various information (eg events, eg EPCIS events) to the event server. These are also transmitted to the assistance system in the form of tax events. Thus, the assistance system not only receives information, for example on the procurement of goods or of raw material for production, but also information about the procurement of the required energy. The assistance system may be implemented distributed in this case, for example, may be arranged a plurality of assistance systems in the energy consumers (company), so that the Steuerereig ^¬ nit be transmitted to at least one assistance system. EPCIS

Exemplary of the EPCIS standard is subsequently used and data exchange, based on the EPCIS standard beschrie ^¬ ben. Here, the functionality of EPCIS is an example extended events to confirm as plans as to also determine from ^¬ deviations from plans.

According to the current standard, an EPCIS event describes which object (what?), At which location (where?), At what time (when?) For what reason (why?) Was observed.

This concept can been enhanced or used ^¬ to represent event events with energy consumption such as Überla ^¬ processing event, tariff change event or Verbrauchsstatus-. For example, it can be expressed that a corresponding tariff (what?) Changes at a certain point in time (when?) For a certain place (where?) For a reason (why?). This enables the detection and notification of problems that occur, for example, along a production (eg a supply chain). The XML-based EPCIS event syntax is based on formal semantics. In particular, the present solution proposes to deduce implicit knowledge from events, which allows situations and compensation strategies to be derived much more efficiently. For EPCIS events and master ontologies refer to IEC 61512 and IEC 62264 standards.

Logical predicates are used in an ontology in order for For (predicates with an arity of two) ^¬ press classes (predicates with an arity of one), and relationships between classes. Statements thus have the form of C (x) or R (x, y), where C be ^¬ draw a class, a relation R and x, concrete objects. Thus, classes represent event types, whereas properties represent event fields.

For example, the statements express

 - Obj ectEvent (e1),

 - Assembly (pl) and

 - bizStep (el, pl)

from that an EPCIS event Obj ectEvent el was observed within a step bizStep pl, which is from the yp assembly. Alternatively, an EPCIS could be used QuantityEvent to the bizstep to step to describe the type Assembly, the standing for Availability checked ^¬ supply electricity. Here, the type epcClass can refer to a resource "stream".

The operators ¥, 3 and the logical links V, Λ, -> ·, etc. can be used to specify complex class descriptions. For example, it can be expressed by the following formula that each individual that is an ObjectEvent event and has a step bizStep that belongs to the Assembly class also an event of a class

ReceivedInAssembly is:

ReceivedInAssembly (x) - - ObjectEvent Cx) Λ

 Λ 3y. (Assembly (y) Λ bizStep (x, y))

New events ObjectEvent x that meet this condition are automatically assigned to ReceivedInAssembly events. All situation detection and compensation rules defined for such ReceivedInAssembly events are automatically applied to the new events. Thus, the number of situation detection and compensation rules can be reduced. This has a positive effect on the specification and the fault tolerance. The situation detection and compensation explained below can be implemented using the EPCIS event ontology or can be extended accordingly.

EVENT PROCESSOR

An event processor may be implemented as part of the assistance system 105. The event processor may respond to the control events 104 based on the information 108 or 109, that is, information or events from the company 101 or the energy supplier 102. In addition to events that originate from production, for example, the event processor can also take into account energy-relevant events in a corresponding manner, and planning can be adjusted based on these input variables. For example, it can be determined by the event processor that a current rate has changed by more than x% be of a Vorga-, which can lead to the production ^¬ tion is adjusted, for example, and / or that the current supplier is gewech ^¬ selt , Consider the following explanations In particular, it is possible to consider the information 109 from the energy supplier 102 for the optimization and / or to generate the feedback 107 to the energy supplier 102 as a result of the optimization ,

It should be noted that the enterprise 101 can include at least one company or at least one technical installation. Accordingly, the energy supplier 102 a

Include a variety of energy suppliers. The feedback 107 may also relate to a selection of an energy supplier.

The goal of the event processor is, for example, to identify critical situations that are based, for example, on timestamped events, current production scheduling, expected inventories, or other events of the energy supplier. As a result of such detection, appropriate countermeasures may be initiated, eg, rescheduling and / or issuing further events, eg so-called EPCIS events. Fig.2 shows exemplarily an Er ^¬ eignis processor, which includes the following components and provides the corresponding functionalities:

 a registration 201,

 a situation detection 202,

 a compensation 203.

These components or functionalities are explained below.

Registration 201

The registry 201 receives and analyzes a producti ^¬ onsplan 204, which was provided for example by the industrial plant (the company 101, for example) in order to register for relevant events. In doing so, the registry 201 updates (see arrow 206) the schedule data in the situation recognition 202. The production plan 204 specifies the resource (for example, energy requirements, materials and end products), which are required to a pre ^¬ given time. For example, production plan 204 fiCPxIxlxT corresponds to a set of tuples (p, m, q, t) describing material usage, where

- p € P a resource identifier (eg material according ei ^¬ ner BOM)

 in G M a machine identifier or a machine location and

 - iy E M is an amount of a resource p that is needed at the time t G T,

describe .

Accordingly, energy-related information 109 can also be detected and taken into account by the energy supplier 102.

FIG. 3 shows an algorithm describing how requests to an event mechanism of an EPCIS database are logged in and how the required availabilities of the situation detection resources 202 are provided.

By way of example, terms from object-oriented programming are used herein to refer to elements of a vector. So denotes an expression xa compo nents a ^¬ of a vector x € X = {(Q, b, c) \ a € € AAHE Ac B C}. The algorithm shown in Figure 3 denotes a procedural ^¬ ren to register relevant EPCIS events at a EPCIS database 205 and to update availability schedules (see. Arrow 206). As input parameters are a Resource consumption plan R and an observation list L provided.

The algorithm maintains an observation list L containing product codes (EPCs) or class identifiers EPCClassIDs of those resources that are already registered. A function for integrating logistics information into product processes represents mapping the internal identifier used for the different resource classes in the material description (MBOM) parts list to the external ones

Identifiers used in the EPCIS tracking and tracing system. This mapping is implemented in line 2 of the algorithm using a so-called lookup EPC method. In the production environment a class-based identification of Materia ^¬ lien may accession game, be used to obtain a production flexibility, since especially in the case of Just-In-Time (JIT) or Just-In-Sequence (JIS) processes a piece-based tracking and Tracing identification is used.

Different configurations of the lookupEPC method can be used to create different scenarios abzude ^¬ CKEN. The difference between internal and external identifiers can be addressed by explicit mapping of EPCClass / EPC identifiers to MBOM (class) identifiers. Based on the structure of the identifier, in many cases the class identifier can be derived directly from the item identifier. For example, the EPC class

 urn: epc: idpat: sgtin: 0614141.112345. *

of the play with the EPC

 urn: epc: idpat: sgtin: 0614141.112345.400

defined by the first part of the EPC.

Once the external EPC or the associated class (EPCClass) was identified, verified, the algorithm determines whether a corresponding call query is re ^¬ gistriert already in the EPCIS database (line 4). If this is not the case, a new query is registered (lines 6 and 8). The EPCIS Standard provides a set of predefined queries (SimpleEventQueries) that are implemented by EPCIS databases. Since situation detection 202 often compares an actual and expected process behavior, material availability plans are updated accordingly. Such Aktu ^¬ alisierungen can for example be represented locally by an event object event with an action "required" to the ontology (line 3) is added.

For a given resource r c-R, the event specifies a time r.i when a material r.p at a location r.m. must be available. For the sake of brevity, the following abbreviation is used to refer to the ObjectEvent event in the EPCIS event ontology:

ObjectEvent (e, t, I, s, b, a, d) = ObjectEvent (e) Λ

 Λ timestamp (e, i) Λ location (e, Ι) Λ

 AbizStep (e, s) Λ businessTransaction (e, b) A Λ action (e, a) Λ disposition (e, d)

in which

 "timestamp" a time or time information,

- "location" a location or location information,

 "bizStep" a step e.g. the production plant or the supply chain,

 - "businessTransaction" a transaction,

 - "action" an action and

 - "disposition" status information

include or denote.

While the identifier e and the time stamp t for all event-nisarten can be used ande ren ^¬ mentioned relations are particularly optional. Properties that are not defined can be noted in the form of a wildcard "·". Situation Detection 202

By way of example situations like a verfrüh ^¬ te provision or late delivery of Zulie- be ferteilen explains machine failures or other disruptions. Any disturbance is, for example, detected by means of at least one BEO ^¬ bach ended event or described.

Accordingly, the situation detection 202 can also take into account faults, situations or events based on information 109 from the energy supplier 102.

In this respect, a description of situations 207 can be used by the situation detection 202.

In event processing, a situation may be determined by inter-event dependencies using event patterns. Event patterns can be viewed as templates that match specific combinations of events. Here is a regelba ^¬ catalyzed event pattern language is used as an example; A rule-based approach has the advantage that a logical formalism can be combined to "reason" over event hierarchies with additional language constructions and temporal reasoning.

An event pattern is constructed from atomic or complex events and extends the logical formalism as follows:

P

IP BIN P1NOT {P). [P, P] ⁽¹⁾ where

 pr is an n-array predicate with arbitrary expressions

 £ i, ..., i "represents,

- q £ R is a non-negative rational number, - WHERE can be used to define constraints that use an expression t and

BIN is a binary operator relating to one of the temporal relations, e.g. in [Allen, J.F. (1983). Maintaining knowledge about temporal in- tervals. Commun. ACM, 26 (11), 832-843].

These temporal operators include

 SEQ, which is a sequence of events,

 AND, indicating that two events are taking place at the same time, and

 OR, which specifies that at least one of two events must occur.

In the present scenario, the predicate pr mostly takes an (atomic) EPCIS event, e.g. an object event.

Situations can be treated as complex events, ie pr (i _{l 5} ... J; ") <- j).

In general, in particular for the case of distributed production networks, three ways of detecting deviations can be distinguished:

(a) the deviation is compared to the local plans of the companies by comparing the actual ^{provision of} resources (event 209) followed by the EPCIS database;

 (b) the company receives a deviation notification e.g. via an EPCIS interpretation event 208 or

(c) actual object tracking information from the EPCIS database is compared locally with expected resource provision information determined by forecasting algorithms. The three approaches are illustrated using the following event patterns according to rule (2) through rule (4). You decide whether to delay incoming resources or outbound products with a unique identifier, according to planned availability and delivery dates.

ActualDelay (e) Λ delay (e. Ί ₂ , t) <-

(D jectEvent {e, i ₂ , L ·, ·, "REQUIRED".) SEQ

 (2) ObjectEventfe, ίχ, l, ·, ·, "OBSERVED", ·)}

Not ifDelay (e) Λ delay (e, ί-ι ^ ι)

(0bjectEvent (e, i ₂ , -, ', "REQUIRED", ^■ ) SEQ

 (3) ObjectEvent (e, t l, ·, ·, "DEVIATION".) ·))

ExpectedDelay (e) Λ delayfe, t <i, t)

(DbjectEvent {e, i ₂ .J _2! ·, ·, "REQUIRED", ·) Λ

ObjectEventfe, t _{l} 1?} ·, ·, "OBSERVED", ·} Λ (4) deli ¥ eryEstimatioti (i ₃ , Ii, l2} Λ t = fi + ts)

 WHERE h <h + h

The rules work for the energy according to e.g. with so-called QuantityEvents (comprising a class with one type).

The rule (2) implements the first approach, by means of encryption ^¬ gleichens time £] to which a resource on a pre ^¬ given place l was observed with time t _2, is added to the needs a resource, according to the production plan at the location , However, it is usually too late to respond adequately if the delay is detected upon receipt of the delivery. For example, an inter-organizational tracking and tracing system offers two alternatives for early entry. The rule (3) replaces the actual observation by a deviation event and thus implements the second approach. Finally, rule (4) uses historical tracking and tracing data from the EPCIS database to determine the approximate delivery time based on the currently available observations. In the example, the forecast function allows iveryEst del imation: LXL- T to estimate the time of arrival / € T based on the ak ^¬ tual position and the destination of a particular commodity. Possible realizations can be considered average delivery periods prior orders, forecast models, neural networks and / or Simu ^¬ lationsmodelle.

In general, time delays in supply chains are just one possible situation of interest to production planning and control. Due to the proposed approach, e.g. using EPCIS event representation and / or the event pattern language, other (critical)

Situations, e.g. Quantity deviations, faulty lines etc. can be detected. Furthermore, it is possible to verify a consistency of defined situations based on the formal semantics of events.

As a result, the situation detection 202 thus provides a recognized critical situation 210 of the compensation 203 for further processing. Compensation 203

As soon as the critical situation 210 has been recorded, notifications and, if appropriate, suitable compensation measures can be initiated. Compensation 203 preferably uses a set of compensation strategies 213 for this purpose

Compensation strategies 213 may be translated into appropriate responses, e.g. EPCIS events 211 or an update of the planning 212 are implemented. An action can take the form of a feedback from the assistance system 105 to the company 101 and / or the energy supplier 102.

The compensation rules have the form comp <- P, where

 P is a situation description according to equation (1) and

- comp is at least one predicate of the rule engine (rule engine) or comprises a combination of several such predicates. Since the number of compensation rules can be very large (several thousand such compensation rules are quite possible), preferably a compact and precise language or description is needed. The lo ^¬ gikbasierte model example used supports this request because event-nishierarchien be used to define rules at different levels of abstraction, which significantly reduces the number of required rules. In addition, the used formal model for consistency control of Kom ^¬ pensationsstrategien can be used.

As an example, consider a compensation processing strategy for detected delays of a supply chain. The logical inference mechanism automatically provides the information be ^¬ riding that the situations Actual Delay (actual delay) NotifiedDelay (determined delay) and Expec- tedDelay (expected delay), which were recognized in the situational recognition 202, all subclasses of the situation Delay (delay) are. This is guaranteed by the following definition:

Delay (.r) - Whether j ectEvent (J-) Λ delay (x. Y) (5)

In the case of electrical energy, the event "Obj ectEvent" can be noted as "QuantityEvent". Based on this equation (5), it is possible to determine an all ^¬ common (and clear) compensation rule for all ^types of delay as follows: notifySclieduler {"time ^" Si, i2) <- delayfe, h, t ₂ ) Λ

1 ookupEPCT ¹ (e, p) ^

Once a deceleration has been detected, the rule triggers the built predicate "notifyScheduler" from that calls the Aktuali ^¬ sierungsverfahren production timing. This process supports updates the resource availability in terms of time and amount and Variegated ^¬ tion of the production maturity of a product.

For a notification of delays, the predicate additionally uses the resource identifier p (eg according to MBOM), the expected time t and the observed / estimated time t, ₂ . In order to map the external EPCIS identifier e to the internal MBOM identifier p, the inverse function of the previously explained lookupEPC method is used.

The following section shows how the production ^¬ running planning responds to the triggered events Planaktua ^¬ capitalization.

REAK IVE PRODUCT ION PLANNING

The production scheduler is also part of the assistance system 105 and determines the detailed production schedule. Here, e.g. the following aspects are considered.

(i) the decision is taken into account, which task (detailed Ma ^¬ schin scheduling) is to be carried out by means of a specific machine.

 (ii) The supply of resources (stock-oriented scheduling), which is particularly susceptible to logistical events (especially in JIT / JIS scenarios), can also be taken into account.

By way of example, the effects of EPCIS events on stock-oriented planning are considered below. account. A detailed machine scheduling with modifiers ^¬ changes over time can be adjusted according to the model integ ^¬ riert. Reactive scheduling algorithm

The aim of the presented reactive Produktionsablaufpla- voltage approach is to calculate the cost-minimizing sequence of Pro ^¬ duktionsjobs J ^p, if the event processor resource deployment plans or updated production deadlines. Production J obs require a set of withdrawal jobs J ^u from a supply. Stocks are replenished by delivery references obs. The available energy can also be understood as a supply.

4 shows a schematic diagram that Depending ^¬ speeds between Produktionsj obs, removal jobs and Lieferj represents obs. Thus, Figure 4 shows a section of a production line with a plurality of production nodes 401, 402, 405 and 409. Each output node has at least one quantity (even be ^¬ is characterized as a supply or inventory) 401, 402, 406, 408 and 410. A product PI is transferred from the inventory 403 of the production node 401 by means of a Ressourcenbereitstellungsauf ^{¬ contract} 411 in the inventory 406 of the production node 405. Accordingly, a product p ₂ from the goods ^¬ 404 consisted of the production node 402 is transferred by means of a delivery ^¬ jobs 412 in the inventory 406 of the production node 405th Means of a removal Jobs 413 and a Ent ^¬ acquisition jobs 414 are the products pi and p ₂ removed from the Warenbe ^¬ stand 406 and processed to a product by means of a production ps 407; the product p ₃ is stored in a Warenbe ^¬ stand 408 by means of a Produktionsj obs 415. Accordingly, the product pj can be transferred by means of another delivery ^¬ jobs 416 in the inventory 410 of the subsequent Pro ^¬ production node 409. A Produktionsj whether j € J ^p is a 3-tuple j (p, d, s), where p € P the MBOM material class identifier of the final product represents ^¬ represents and d £ N is the due date, wherein both of the Pro ^¬ production orders received, for example, via a production manager. An optimal start time s £ N of obs Produktionsj example, is calculated by a Produktionsab ^¬ run planner. As soon as a function notifyScheduler ("duedate" .p, ii. I ₂ ) with the parameters ("duedate", p,, i ₂ ) is called by the event processor, the production jobs are updated as follows: J ^p = J ^p \ (p,, s) and

J ^p = J ^p U fa t ₂ , s)

Basically updates can resolve the below explained their rescheduling algorithm (Umplanungs algorithm) from ^¬.

Each production year requires a (non-empty) set of resources (products, materials, etc.) from inventories. Production scheduling therefore requires information about resource deployments J °. It is described by the tuples (p, q, t), where p 6 P denotes the MBOM identifier, q EN the quantity and £ N of the delivery date. During the scheduling of the set of Lieferj obs J ^D is used to verify whether a set of sampling job or Verbrauchsj obs is possible at a given time. The set of resource provisioning is made by means of the methods notii ^' yScheduler ("quantity" ./). li) and notifySc edulerC'time ";; ii -'i) updated, as described above to the Produktionsj obs. The Entnahmej obs that will lead to one certain whether Produktionsj be ^¬ can mborn using the function: P -42 ^{FXZ ™} be determined by which the quantity of each required for a ^¬ be-determined product resource is specified. The entire set of sampling jobs for the product p _e "d € P is carried = {(p, q ,. t) \ (p, q) € rnbom (j) _enx i) A ie N} determined where t is the sampling time the resources for the product j _crff | defined and depends on the flow chart of the production ^¬ jobs, see equation (9).

Rescheduling Algorithm (Rescheduling Algorithm):

For example, the Rescheduling algorithm may determine an optimal or optimized flowchart of the production j obs contained in J ^p .

Thus, a flowchart defines a starting point js for all j ζ ^r . A flowchart is thus an overall function σ: J ^p -5> N, which assigns a start time to each job. The goal of selecting the function σ is to minimize a cost function. The cost function may include various cost categories. As an example, a premature or delayed by the distance to a due date

is determined, or even a required amount of a resource (eg energy) serve. Preferably, as part of the rescheduling, a change in the flowchart σ _{0Μ is} only made if this _{results in} a significant improvement being achieved. To quantify the improvement, the distance between the optimal and the current schedule is calculated using a distance function edit.

This function edit is preferably selected depending on the stock ^¬ system. For example, in the case of a high-bay warehouse ^¬ any exchange between Produktionsj obs can be carried out; In this case, the edit function can be implemented using the Hamming distance. Thus, the optimization problem for choosing a best-fitting schedule can be formulated as follows: arg min w · \ & () - jd \ + (1- ^w ) · edit (a, a "u)

a-jP-tN Ί, ^{* "} where w 6 [0, 1] represents a" rescheduling threshold "(vi = l always chooses the optimal schedule, whereas w = 0 leads to the minimal change schedule).

In principle, different variants of a one-dimensional or multi-dimensional optimization problem can be formulated. For the solution, different approaches, e.g. Optimization of a cost function, search of a pareto-optimal solution, etc. can be used.

To ensure that no Produktionsj obs planned who, ^¬ without

 - sufficient stocks or resources,

 - enough time to transport components from the warehouse to the machines, and

- sufficient production capacity are available, for example, the following restrictions according to the formulas (8) to (10) apply. A function defines the prep Trans ^¬ port for predetermined time sampling and Produktionsj obs, a PROD constant-C AP ACity defines how many production ^¬ jobs can be executed in parallel.

V € J ^F , V € yyyy: j'-i - prep (j ') = js (9) i <= r: Ii? GJ ^p Ii-s = t} l <PROD.CÄPA CITY (10)

Equation (8) takes into account that there are sufficient resources available, that equation (9) takes into account that the time for transport is sufficient, and equation (10) ensures that the production capacity is sufficient.

The optimization problem can be used as a mixed integer linear program presented with a fully unimodular Be ^¬ schränkungsmatrix. Thus, the problem can be solved efficiently using a simplex algorithm.

Thus, it is particularly proposed to operate an industrial An ^¬ location, such as a production or logistics system optimized, whereby the power consumption with respect to a

Target function is optimized. In addition to the power consumption, the objective function can also take into account further parameters, for example of an energy supplier or the installation itself, so that, for example, a multi-objective optimization can be carried out and the installation can be adapted with regard to the power consumption or the power consumption with regard to the installation. Both the industrial plant and a power supplier provide information that can be taken into account for optimization ^¬ tion or as part of the objective function. It is advantageous that overload situations are avoided and in particular a large number of regenera ^¬ tive energy sources can be used as energy suppliers, because an adjustment to the actual amount of energy provided and thus the power grid accordingly can be operated or charged. The invention may for example be used in smart grids or in production or lodging ^¬ policy management systems. Although the invention in detail by a ge ^¬ showed embodiment has been illustrated and described in detail at least, so the invention is not limited to this and other variations can be derived therefrom by the skilled artisan without departing from the scope of the invention.

Abbreviations:

EPC Electronic Product Code - Electronic product code

EPCIS Electronic Product Code Information Services - EPC

 information service

ERP Enterprise Resource Planning (system for producti ^¬ onsplanung)

IT information technology

 JIS Just-In-Sequence - timely with regard to, for example, a further processing step (sequence synchronous

Production)

 JIT Just-In-Time - in time with respect to, for example, one

Further processing step (requires synchronous production ^¬ tion)

MBOM Manufacturing Bill Of Material - Parts List

MES Manufacturing Execution System (process-level operating a multi-layered manufacturing management system ^¬)

 RFID Radio Frequency Identification (Identification with

 Help of electromagnetic waves)

 SCM Supply Chain Management

Claims

claims

1. Method for setting an industrial plant,

- in which a power consumption of the industrial plant is determined,

- in which an optimization of power consumption is carried out in ^¬ industrial plant with respect to a target function,

- in which the industrial system is set according to the Op ^¬ optimization.

2. The method of claim 1, wherein the objective function information about past power consumption be ^{¬ taken into} account.

3. The method according to any one of the preceding claims, wherein the target function information from an energy supplier, in particular a signal regarding the power consumption, taken into account.

4. The method of claim 3, wherein the information from the energy supplier comprises at least one of the following information:

 a signal relating to the minimum and / or maximum power consumption to be set;

 - a price information;

 - another fare information;

 - a utilization information.

5. The method according to any one of the preceding claims, wherein the objective function information of the power grid, in particular a state of the power grid, taken into account ^¬ taken.

6. The method according to any one of the preceding claims, wherein the objective function takes into account a current, a past and / or a planned power consumption of the industrial plant.

7. The method according to any one of the preceding claims, wherein the optimization of the power consumption and at least one further parameter of the industrial plant is performed with respect to the objective function.

8. The method of claim 7, wherein the at least one further parameter comprises:

 - a parameter of the industrial plant;

 - a lead time;

 - costs of the system;

 - cost of electricity;

 - a logistics plan;

 - a production plan.

9. The method according to any one of the preceding claims, wherein the objective function takes into account a current or a future situation of the industrial plant.

10. The method of claim 9, wherein the current or future situation of the industrial plant takes into account at least one of the following information:

- a delivery status of supplied or ready frame ^¬ ter goods;

 - an expansion status of the industrial plant;

 - an operating mode of the industrial plant.

11. The method according to any one of the preceding claims, wherein the objective function comprises at least one of the following objectives:

 - compliance with a delivery commitment;

 - compliance with a given machine utilization;

- compliance with a given energy consumption;

 - compliance with a throughput.

12. The method according to any one of the preceding claims, ^wherein the industrial plant comprises at least one of the following ^¬ components: - a production or manufacturing;

 - a logistics unit;

 - a service unit;

 - a data center;

 - an energy supplier;

 - a transmission system operator.

13. An apparatus for adjusting an industrial plant is comprising a processing unit which ^¬ is oriented such that

 a power consumption of the industrial plant can be determined,

- optimizing the power consumption of industriel ^¬ len plant is feasible in terms of an objective function,

 - The industrial plant is adjustable according to the optimization.

14. The apparatus of claim 13, which is part of a production management system.

15. System comprising at least one device according to one of claims 13 or 14.