Energy-optimized operation of modular industrial plants

Abstract

A method (100) for operating an industrial plant (1) that is configured to execute an industrial process using a plurality of interconnected process modules or other functional units (2a-2f), wherein at least one process module or other functional unit (2a-2f) is capable of performing its assigned role in the industrial process in different modes of operation (3a-3d) that have different energy consumptions, and wherein the method (100) comprises the steps of: - obtaining (110) energy information (4) that is indicative of the availability, and / or of the cost, of energy required for the operation of at least one process module or other functional unit (2a-2f); - obtaining (120), for at least one chosen process module or other functional unit (2*, 2a-2f), a candidate mode of operation (5) that is desirable from the point of view of this process module or other functional unit (2*, 2a-2f) with respect to energy consumption; - determining (130) whether operation of the chosen process module or other functional unit (2a-2f) if this candidate mode of operation (5) is compatible with the intended operation of one or more further process modules or other functional units (2a-2f) that are connected to the chosen process module (2*, 2a-2f); and - if this is the case, authorizing (140), in an energy plan (1a) for operating the industrial plant (1), operation of the chosen process module or other functional unit (2*, 2a-2f) in the candidate mode of operation (5).

Description

[0001] ABB Schweiz AG 07.10.2024

[0002] A 19307 WO

[0003] ENERGY-OPTIMIZED OPERATION OF MODULAR INDUSTRIAL PLANTS

[0004] FIELD OF THE INVENTION

[0005] The invention relates to the operation of industrial plants where a plurality of process modules or other functional units work together to execute an industrial process.

[0006] BACKGROUND

[0007] Sustainability is one of the largest topics of today. A large portion of the CO2 equivalent pollution is coming from process plants of different industries, such as chemical, metals, or pharmaceutical. Therefore, industry is going through a transition for a more sustainable and energy efficient operation.

[0008] One cornerstone of this is electrification. Process heat, as an example, is not produced by burning natural gas, but using electricity. Here the assumption is that electricity will be produced using more renewables in future and thus, being sustainable. A second, similarly important, cornerstone is saving energy. This is especially needed during times when less electricity can be produced using renewables, and thus more electricity must be produced using coal or gas fired power plants.

[0009] Taking this into account, the present invention proposes a method to optimize the energy consumption of modular engineered plants. It describes a concept of environmental aware modules that can optimize themselves by switching energyprocedures automatically if needed and communicate this switch to other modules for further reaction.

[0010] Today’s process plants can usually not simply switch into more energy efficient operation modes if needed. They are tuned to a certain steady state which produces the needed amount of output and leaving this state is not trivial. This means, a reaction to changed environmental aspects, such as changed energy mix, energy price, P240519W001 - 2 - 07.10.2024 weather is simply not possible or a very slow - often too slow - process, since it must be incorporated manually into the stiff processes.

[0011] OBJECTIVE OF THE INVENTION

[0012] It is therefore an objective of the present invention to facilitate the operation of an industrial plant in a manner that is optimized with respect to energy usage.

[0013] This objective is achieved by a method according to the independent claim. Further advantageous embodiments are detailed in the dependent claims.

[0014] DISCLOSURE OF THE INVENTION

[0015] The invention provides a method for operating an industrial plant. This industrial plant is configured to execute an industrial process. To this end, a plurality of interconnected process modules or other functional units work together.

[0016] In particular, process modules may be self-contained units that

[0017] • have one or more input ports for one or more commodities and / or energy,

[0018] • have one or more output ports for one or more commodities and / or energy, and

[0019] • provide a particular service towards the execution of the industrial process.

[0020] For example, the process module may take in one or more chemical substances, heat, stir, react or otherwise process the one or more chemical substances, and then deliver the result of this processing as output. Such process modules may be linked in a temporary manner to form an industrial plant executing a particular process, and then un-linked an re-assembled like Lego to form an industrial plant that executes a different process.

[0021] But in the context of the present invention, it is not required that the industrial plant is composed of process modules that are so self-contained. Rather, it suffices that the industrial plant comprises interconnected functional units, i.e., some division of the industrial plant into such interconnected functional units is discernible.

[0022] At least one process module or other functional unit is capable of performing its assigned role in different modes of operation that have different energy consumptions. In particular, in this manner, it may be possible to select a trade-off between operating speed or throughput on the one hand, and low energy usage on the other hand. Such P240519W001 - 3 - 07.10.2024 functionality is present even in household appliances, such as washing machines, dryers or dishwashers. Usually, an energy-saving cycle will achieve the same result as a standard cycle, but it will use less energy, and in exchange for this take much longer to complete the cycle. For example, in the washing machine or the dishwasher, the savings is brought about by re-cycling one and the same amount of water more often, rather than replacing it with fresh water and heating this.

[0023] In this example, the energy is electrical energy. But energy can take all suitable forms. For example, energy may also be provided in the form of a fuel, such as natural gas, hydrogen, oil or an oil product, or in the form of compressed air, water in an elevated reservoir, or any other form of mechanically stored energy.

[0024] In the course of the method, energy information is obtained. This energy information is indicative of the availability, and / or of the cost, of energy required for the operation of at least one process module or other functional unit. For example, the energy information may comprise information about a current or future energy mix, or a current or future energy price. But since the production of renewable energy is largely weatherdependent, the energy information may also comprise the current weather or a weather forecast. One example of energy information is the spot price for electricity that is set at the EPEX day-ahead auction for every hour of the next day and that is already being used to set the price for many “flexible” household and industrial customers.

[0025] For at least one chosen process module of other functional unit, a candidate mode of operation that is desirable from the point of view of this process module or other functional unit with respect to energy consumption is determined. In the example of the washing machine, dryer or dishwasher, this may, for example, be the designated energy-saving cycle.

[0026] It is then determined whether operation of the chosen process module or other functional unit in this candidate mode of operation is compatible with the intended operation of one or more further process modules or other functional units that are connected to the chosen process module. If this is the case, in an energy plan for operating the industrial plant, operation of the chosen process module or functional unit in the candidate mode of operation is authorized. In this process, in particular, only direct, nearest neighbours of the chosen process module or other functional unit in P240519W001 - 4 - 07.10.2024 terms of the topology of the plant may be considered. That is, if the chosen module A is connected to module B, and then module B is connected to module C, only the interaction between modules A and B is considered.

[0027] In this manner, the interdependencies between the process modules or other functional units that work together to execute the given industrial process as a whole may be taken into account in a much simpler manner than by regarding the industrial process as a monolithic entity and trying to optimize the many degrees of freedom that this industrial process has all at once. It is well possible that, by regarding only one process module or functional unit and interactions with its respective nearest neighbours at a time, a global optimum that could have been obtained by optimizing the process as a whole with all its degrees of freedom may be missed. But this optimization may just take too long for any practical purposes, so, for example, by the time the computation is finished, a “golden hour” of particularly low energy prices may long be over.

[0028] That is, a centralized model / control may just require too much model information, which may lead to a too complicated problem to solve. The overcomplicated model may lead to long computational time. A mere simplification of the model, on the other hand, may lead to the results / solved strategy not being available or optimal. Besides, the high- level supervisor system may not have the functionality of doing centralized optimization / control. Thus, in conventional plants that are not aware of the environment and that are not engineered for rapidly switching into modes, it is less likely to save energy during operation.

[0029] In a toy example from the household, the industrial process is getting dirty clothing ready for use again, which requires the use of the washing machine as the first module and subsequent use of the dryer as the second module. If the washing machine runs in the energy-saving cycle, the dryer may have to wait longer before it has something to dry, and if the dryer itself runs in its energy-saving cycle as well, the total process may take too long. A window of low electricity prices, which may, e.g., comprise a few daytime hours around noon with a lot of solar power and not too much electricity demand, may then already be over. The savings may then be greater if the washing machine runs a normal, faster cycle that uses more energy, so that the dryer can consume its far larger consumption in the cheapest hours. P240519W001 - 5 - 07.10.2024

[0030] In a particularly advantageous embodiment, after it has been determined in which mode of operation the chosen process module or other functional unit shall be operated, a next process module or other functional unit may be chosen, and the method may be continued again from the obtaining of a candidate mode of operation for this new chosen process module or other functional unit. In this manner, the optimization of energy usage may propagate throughout the industrial plant, from one process module or other functional unit to the next. In particular, this iterative process may continue until all process modules or other functional units of the industrial plant have been “touched”. That is, energy usage of the industrial process as a whole is optimized to some degree, but without the complexity of having to consider all degrees of freedom at once.

[0031] It is not guaranteed that, for every chosen process module or other functional unit, a candidate mode of operation will be authorized in the energy plan. For example, as in the above toy example, a constraint may be in the way in that running all process modules at minimum energy usage may slow the process down too much. Therefore, in a further particularly advantageous embodiment, for each process module or other functional unit for which operation in a candidate mode of operation has not been authorized, the energy plan for operating the industrial plant comprises operation of this process module or other functional unit in a predetermined mode according to the engineering of the industrial plant. That is, the optimization may start from a configuration of modes of operation that is known to be fit for executing the industrial process by virtue of the engineering, and depart from this initial configuration only to the extent that this optimizes energy usage.

[0032] There are multiple possibilities for choosing the process module or other functional unit whose candidate mode of operation is to be changed. For example, one may choose the first, i.e. , the most upstream, process module or other functional unit in the topology of the industrial plant. For example, in a chemical process, this may be the process module or other functional unit that receives the initial educts of the chemical process. The optimization may then propagate along the progress of the chemical process from one process module or other functional unit to the next. In another particularly advantageous embodiment, a process module or functional unit with a highest energy consumption is chosen as the chosen process module or functional unit. In this P240519W001 - 6 - 07.10.2024 manner, the biggest possible savings may be locked in first, and then other savings may be realized optionally.

[0033] In a further particularly advantageous embodiment, in response to determining that a predetermined criterion for the finishing of the energy plan is met, operation of the industrial plant according to the energy plan is caused. In this manner, the energy savings that are promised by the energy plan are put into practice in the actual operation of the industrial plant. Causing the operation according to the energy plan may, for example, comprise communicating, among process modules or other functional units, and / or between a central control entity of the industrial plant on the one hand and the process modules or other functional units on the other hand, that the respective process module or other functional unit shall start operating in its respective operating mode as per the energy plan starting now, or starting at a later time. Switching to the new plan as of now is more advantageous for continuous industrial processes, whereas switching to the new plan effective at a later time is more advantageous for batch processes that produce a batch of product at a time.

[0034] There is complete freedom regarding the criterion when the energy plan is to be considered finished. In one example, the criterion may comprise that all process modules or other functional units of the industrial plants have been chosen process module or functional unit in the course of the method. That is, each process module or other functional unit has been “touched” at least once, and an attempt has been made to find a suitable mode of operation for it. Optionally, the optimization may continue for one or more further such passes through the industrial plant. It is possible that, after the modes of operation of process modules or other functional units further downstream have been optimized, this creates new room for optimizing the mode of operation of earlier, further upstream process modules or other functional units again. In another example, the criterion may comprise that operating of the industrial plant according to the energy plan causes a predetermined goal in terms of energy and / or cost savings to be met. Using this energy plan, even though the global optimum regarding energy usage may not yet have been reached, may be more economic than “hunting” further for the global optimum.

[0035] In a further particularly advantageous embodiment, the desirable candidate mode of operation is determined by the chosen process module or functional unit. As initially P240519W001 - 7 - 07.10.2024 noted, one difficulty in optimizing the energy usage in a complex industrial plant is that a central control entity may not be aware of all the details of all the process modules or other functional entities that may impact the possibilities to save energy. Even though the central control entity may be aware of the type of each process module or other functional unit, the possibilities for switching modes of operation may depend on the concrete operating states of the chosen process module or other functional unit, and on the concrete operating states of neighbouring process modules or functional units in the topology of the plant. The central control entity may not be aware of all these operating states. Therefore, decisions on candidate modes of operation that are made locally by the chosen process module or functional unit are more founded than decisions made by a central control entity. This is in some way analogous to disaster control where most decisions have to be made locally in the affected area, rather than by a central government far away.

[0036] In a further particularly advantageous embodiment, the determining whether operation of the chosen process module or other functional unit in the candidate mode of operation is compatible with the intended operation of one or more further process modules or other functional units that are connected to the chosen process module comprises a negotiation between the chosen process module or other functional unit and the further process modules or other functional units. That is, the chosen process module or other functional unit may ask a neighbouring process module or other functional unit whether the switch to the intended mode of operation will be acceptable from the perspective of this neighbouring process module or other functional unit, or whether the operation of this neighbouring process module or other functional unit may be impeded. For example, a feeder module may ask a downstream reactor module a question of the kind, “May I switch to my energy-saving mode that delivers only 50 % of the usual output at 30 % of the energy usage?” The reactor module may then reply, “OK, no problem with that, I’m presently full with a reacting batch of educts anyway”, or “No, sorry, at the moment I need every bit of this educt that I can get hold of because there is a lack of it in the currently reacting batch.”

[0037] Thus, in a further particularly advantageous embodiment, in the course of the negotiation, at least one constraint by virtue of which operation of the chosen process module or other functional unit in the candidate mode of operation is not compatible with the intended operation of the further process module or other functional unit is P240519W001 - 8 - 07.10.2024 communicated from the further process module or other functional unit to the chosen process module or other functional unit. For example, such a constraint may relate to a quantitative amount or rate of input that needs to be delivered by the chosen process module or other functional unit to the further process module or other functional unit so that this further process module or other functional unit can meet is own obligations in the context of the industrial process. For example, the reactor module may tell each feeder module that it needs at least a certain rate in liters per minute of each educt, and at a respective minimum temperature. The chosen process module or other functional unit may then determine a new candidate mode of operation given the at least one constraint. In the example of the reactor and the feeders, it may not be possible to switch a feeder to intermittent operation without causing too little educt to be delivered to the reactor. But it may be possible to at least set a pump in the feeder to a slower rotation speed to conserve some energy.

[0038] In a further particularly advantageous embodiment, energy information is provided to the chosen process module or other functional unit by a central control entity of the industrial plant. As discussed before, the central control entity may not be aware of concrete operating states of each and every process module or other functional unit. But it may be aware of facts that affect the industrial plant as a whole, such as the weather, or energy prices. Also, it may depend on the intricacies of each individual process module or other functional unit how its operation and energy usage are affected by a particular fact in the energy information. For example, if the ambient air temperature falls, this may translate to an increased energy usage in a module that needs heating, but to a lower energy usage in a module that needs cooling.

[0039] In a further particularly advantageous embodiment, a central control entity of the industrial plant communicates at least one production constraint to the chosen process module or functional unit. The chosen process module or functional unit determines the desirable candidate mode of operation given this production constraint. The chosen process module or functional unit itself is better informed than the central control entity regarding how the production constraint translates to a constraint of the mode of operation. Considering the production constraint right from the start saves time and also yields a potentially better optimization result than first optimizing without this constraint and then checking whether the constraint is met. P240519W001 - 9 - 07.10.2024

[0040] The determining whether operation of the chosen process module or other functional unit in this candidate mode of operation is compatible with the intended operation of one or more further process modules or other functional units that are connected to the chosen process module may have many facets. As discussed before, this determining may, for example, be made dependent on whether a further process module or other functional unit that is connected downstream of the chosen process module or other functional unit is able to deliver its required contribution to the industrial process despite a reduced output rate of the chosen process module or other functional unit. Alternatively or in combination to this, the decision may be made dependent on whether a goal of the industrial process as a whole can still be met despite operation of the chosen process module or other functional unit in the candidate mode of operation. Alternatively or in combination to this, as another criterion, it may be checked whether a further process module or other functional unit that is connected upstream of the chosen process module or other functional unit is able to deliver its output despite a reduced intake rate of this output by the chosen process module or other functional unit. On the face of it, this reduced intake rate is not critical, but if this leads to the product of an upstream process module or other functional unit having nowhere to go, this may impede the operation of this upstream process module or other functional unit. This upstream process module or other functional unit may have to slow its rate of production as well, if this is possible without causing trouble elsewhere in the industrial process.

[0041] In particular, a process module or other functional unit that delivers a final product of the industrial process may determine whether the goal of the industrial process as a whole can still be met. In many use cases, the goal of the industrial process as a whole is formulated in terms of such a quantitative requirement.

[0042] When considering such quantitative requirements, buffer capacities may advantageously be taken into account as well. A process module or other functional unit that needs a particular educt may have some buffer capacity for this educt, so that it does not immediately run out of this educt in case there is a temporary hiccup in the supply. Likewise, a process module or other functional unit that produces some product may have some buffer capacity for this product that ensures that the product always has somewhere to go. Therefore, in a further particularly advantageous embodiment, whether a further process module or other functional unit that is connected downstream P240519W001 - 10 - 07.10.2024 of the chosen process module or other functional unit is able to deliver its required contribution to the industrial process despite a reduced output rate of the chosen process module or other functional unit is made dependent on whether the further process module or other functional unit has a sufficient stock of the output of the chosen process module or other functional unit. Alternatively or in combination to this, whether a further process module or other functional unit that is connected upstream of the chosen process module or other functional unit is able to deliver its output despite a reduced intake rate of this output by the chosen process module or other functional unit is made dependent on whether there is sufficient buffer capacity for accepting the output of the further process module or other functional unit.

[0043] In this context, it is particularly advantageous to determine the energy plan for a predetermined period of time in the future. First, the energy information, such as weather forecasts and energy prices, have a limited temporal validity. Second, the mentioned buffer capacities are finite and can therefore help out only for a limited amount of time if there is too much or too little of some commodity. Thus, in an otherwise identical situation, whether operation of the chosen process module or other functional unit in the desired candidate mode of operation is possible or not may depend on the length of the period of time for which the energy plan is computed.

[0044] In a further particularly advantageous embodiment, multiple candidate energy plans are determined starting from different chosen process modules or other functional units. A candidate energy plan that is optimal with respect to a given optimization goal is then chosen as the final energy plan. In this manner, a good solution may also be found in situations where starting with the most upstream process module or functional unit, or the biggest energy consuming process module or functional unit, is not optimal. For example, if the biggest energy consumer is chosen, this may prevent a lot of other process modules or other functional units from entering respective energy-saving modes of operation and causing larger overall savings than the biggest energy consumer alone could accomplish. In particular, the testing of multiple candidate energy plans may be performed using a “digital twin” of the industrial plant in one or more computers. This can be done faster than in real time and without overly taxing the compute capabilities of the process modules or other functional units. In particular, when working with such a “digital twin”, the propagation of the optimization throughout P240519W001 - 11 - 07.10.2024 the plant by negotiating between neighbouring process modules or other functional units may be emulated.

[0045] Because it is computer-implemented, the present method may be embodied in the form of a software. The invention therefore also relates to a computer program with machine-readable instructions that, when executed by one or more computers and / or compute instances, cause the one or more computers and / or compute instances to perform the method described above. Examples for compute instances include virtual machines, containers or serverless execution environments in a cloud. The invention also relates to a machine-readable data carrier and / or a download product with the computer program. A download product is a digital product with the computer program that may, e.g., be sold in an online shop for immediate fulfilment and download to one or more computers. The invention also relates to one or more compute instances with the computer program, and / or with the machine-readable data carrier and / or download product.

[0046] DESCRIPTION OF THE FIGURES

[0047] In the following, the invention is described using Figures without any intention to limit the scope of the invention. The Figures show:

[0048] Figure 1 : Exemplary embodiment of the method 100 for operating an industrial plant;

[0049] Figure 2: Exemplary industrial plant 1 with interlinked process modules 2a-2d and a central control entity 6 that provides energy information 4 to all modules 2a-2d;

[0050] Figure 3: Exemplary communication between a chosen process module 2*, 2a, a further downstream process module 2b, and the central control entity 6.

[0051] Figure 1 is a schematic flow chart of an embodiment of the method 100 for operating an industrial plant 1. The industrial plant 1 is configured to execute an industrial process using a plurality of interconnected process modules or other functional units 2a-2f. At least one process module or other functional unit 2a-2f is capable of performing its assigned role in the industrial process in different modes of operation 3a- 3d that have different energy consumptions. P240519W001 - 12 - 07.10.2024

[0052] In step 110, energy information 4 that is indicative of the availability, and / or of the cost, of energy required for the operation of at least one process module or other functional unit 2a-2f is obtained. For example, this may be determined by a central control entity 6 of the industrial plant 1 and communicated to the process modules or other functional units 2a-2f.

[0053] In step 120, for at least one chosen process module or other functional unit (2*, 2a-2f), a candidate mode of operation (5) that is desirable from the point of view of this process module or other functional unit (2*, 2a-2f) with respect to energy consumption is obtained.

[0054] According to block 121 , a process module or functional unit 2a-2f with a highest energy consumption may be chosen as the chosen process module or functional unit 2*.

[0055] According to block 122, the desirable candidate mode of operation 5 may be determined by the chosen process module or functional unit 2*, 2a-2f. For example, this may be done based on energy information 4 received from the central control entity 6 of the industrial plant 1.

[0056] According to block 123, a central control entity 6 of the industrial plant 1 may communicate at least one production constraint to the chosen process module or functional unit 2*, 2a-2f. According to block 124, the chosen process module or functional unit 2*, 2a-2f may then determine the desirable candidate mode of operation 5 given this production constraint.

[0057] In step 130, it is determined whether operation of the chosen process module or other functional unit 2a-2f if the candidate mode of operation 5 is compatible with the intended operation of one or more further process modules or other functional units 2a- 2f that are connected to the chosen process module 2*, 2a-2f. If this is the case (truth value 1), in an energy plan 1a for operating the industrial plant 1 , operation of the chosen process module or other functional unit 2*, 2a-2f in the candidate mode of operation 5 is authorized. P240519W001 - 13 - 07.10.2024

[0058] According to block 131, the determining whether operation of the chosen process module or other functional unit 2*, 2a-2f in the candidate mode of operation 5 is compatible with the intended operation of one or more further process modules or other functional units 2a-2f that are connected to the chosen process module 2*, 2a-2f may comprise a negotiation between the chosen process module or other functional unit 2*, 2a-2f and the further process modules or other functional units 2a-2f.

[0059] In the course of this negotiation, according to block 131a, the further process module or other functional unit 2a-2f may communicate, to the chosen process module or other functional unit 2*, 2a-2f, at least one constraint by virtue of which operation of the chosen process module or other functional unit 2*, 2a-2f in the candidate mode of operation 5 is not compatible with the intended operation of the further process module or other functional unit 2a-2f. That is, the chosen process module or other functional unit 2*, 2a-2f learns about the exact problem that precludes entry into the candidate mode of operation 5. The chosen process module or other functional unit 2*, 2a-2f may then, according to block 131b, determine a new candidate mode of operation 5 given the at least one constraint.

[0060] According to block 132, it may be determined whether a further process module or other functional unit 2a-2f that is connected downstream of the chosen process module or other functional unit 2a-2f is able to deliver its required contribution to the industrial process despite a reduced output rate of the chosen process module or other functional unit 2*, 2a-2f. To this end, in particular, according to block 132a, it may be checked whether the further process module or other functional unit has a sufficient stock of the output of the chosen process module or other functional unit 2*, 2a-2f.

[0061] According to block 133, it may be determined whether a further process module or other functional unit 2a-2f that is connected upstream of the chosen process module or other functional unit 2*, 2a-2f is able to deliver its output despite a reduced intake rate of this output by the chosen process module or other functional unit 2*, 2a-2f. To this end, in particular, according to block 133a, it may be checked whether there is sufficient buffer capacity for accepting the output of the further process module or other functional unit 2*, 2a-2f. P240519W001 - 14 - 07.10.2024

[0062] According to block 134, it may be determined whether a goal of the industrial process as a whole can still be met despite operation of the chosen process module or other functional unit 2*, 2a-2f in the candidate mode of operation 5. To this end, according to block 134a, this determining may be made by a process module or other functional unit 2a-2f that delivers a final product of the industrial process.

[0063] In the example shown in Figure 1 , in step 150, a candidate mode of operation 5 is determined for a next chosen process module or other functional unit 2*, 2a-2f. That is, the method branches back to step 120 with a new chosen process module or other functional unit 2*, 2a-2f, so as to augment the energy plan 1a.

[0064] In the example shown in Figure 1 , in step 160, it is checked whether a predetermined criterion for the finishing of the energy plan 1a is met. If this is the case (truth value 1), in step 170, operation of the industrial plant 1 according to the energy plan 1a is caused. That is, the energy plan 1a is applied to the industrial plant 1.

[0065] Figure 2 illustrates an exemplary industrial plant 1 with four process modules 2a-2d, namely a feeder 2a, a reactor 2b, a heat exchanger 2c, and a distillation 2d. The feeder 2a takes in the educts E of the industrial process and forward them to the reactor 2b. The reactor 2b reacts the educts and, to this end, calls upon a heat exchanger 2c to heat them up. The product obtained by the reaction is then forwarded to a distillation column 2d that then outputs the final product P of the industrial process.

[0066] In a known modular architecture, the modules 2a-2d simply receive commands (labelled C) from the central control entity 6 to start and stop services, and the modules 2a-2d do not communicate to each other. All communication is done through the central control entity 6: The central control entity (also known as process orchestration layer, POL) 6 issues, to modules 2a-2d, commands C to start and stop services, and receives, from modules 2a-2d, information about their respective service states S. Thus, in standard modular engineered plants, only the communication of commands and states (green arrows) is used. The information about the topology, as indicated by the horizontal arrows between modules 2a-2d, is known to the central control entity 6, but not to the modules 2a-2d themselves. Compared with the prior art, in the example shown in Figure 2, the present method adds, in the communication between the modules 2a-2d on the one hand and the central control entity 6 on the other hand, P240519W001 - 15 - 07.10.2024

[0067] • communication of the energy information 4 from the central control entity 6 to the modules 2a-2d, as well as

[0068] • communication of the currently executed service procedure, SP, from the process execution assemblies, PEA, in the modules 2a-2d to the central control entity 6.

[0069] In particular, the central control entity 6 is used to gather information about the environmental context, which means the currently used energy mix, the energy price, the weather forecast (to predict the future energy mix and price) and other relevant items. This context is communicated to the modules 2a-2d, as well, to make them environment aware.

[0070] Additionally, the modules 2a-2d know their process context. Every module 2a-2d is connected during engineering to other modules 2a-2d. This information can be reused to tell every module 2a-2d which are the preceding and succeeding modules 2a-2d. That means, the modules 2a-2d also know their context and “speaking partners” and can later communicate with them. In Figure 2, the process connections are indicated using horizontal arrows between modules 2a-2d. Previously, these connections were only known to the central control entity 6, but not to the modules 2a-2d.

[0071] Furthermore, the modules 2a-2d are engineered using several different energyprocedures 3a-3d for every service provided by them. The energy-procedures 3a-3d allow doing the same thing but have different energy consumption and output. The energy consumption and the corresponding output as well as possible side-effects like wear-down related to the procedure or buffer-capacities (buffer can be related to material (having a tank) or to energy (e.g. with heat)) are known by the modules 2a-2d themselves, as well. Switching between those procedures is triggered by the module 2a-2d itself.

[0072] The circles in the modules 2a-2d in Figure 2 indicate an intelligence in the modules 2a- 2d. This intelligence holds the information about the service (incl. energy consumption), strategies for changing the energy-procedures (modes of operation) 3a-3d, the connection to the process context and the current environmental context as delivered by the central control entity 6. P240519W001 - 16 - 07.10.2024

[0073] Figure 3 illustrates the communication between the feeder 2a, the reactor 2b, and the central control entity 6 of the industrial plant 1 shown in Figure 2 in more detail.

[0074] In step a, the central control entity 6 communicates the environmental context as energy information 4 to the modules 2a and 2b. The context includes the current energy mix, the current energy price, the weather forecast and perhaps other values of interest. Furthermore, the central control entity 6 communicates critical production constraints, such as needed volume for the production or maximum volume due to storage constraints. This information is used by the modules to evaluate which energyprocedure, i.e., which mode of operation 3a-3d, fits best to the current context, later. An energy-procedure might be implemented like a module type package, MTP, standard compliant “procedure”, or as a sub-procedure of the same.

[0075] In step b, since the central control entity 6 contains the recipe or sequence to be executed and “knows” the process, it now sends commands C to the modules 2a, 2b for executing a fitting service. It does not send an energy-procedure 3a-3d to be executed, because that is decided by the module 2a, 2b itself. It only sends the command C to start a service.

[0076] Based on steps a and b, the chosen module 2a, 2* (here: the feeder) starts evaluating which energy-procedure of the requested service shall be executed, i.e., which mode of operation 3a-3d should be entered.

[0077] The receiving module 2a uses the information from the POL, and the information about which service must be started, to evaluate which energy-procedure 3a-3d of the service fits best to the production constraints, with the least (energy) cost needed. Furthermore, it can evaluate when to switch into an energy-procedure with higher energy consumption, but with more throughput, in case there is a good energy mix or price at present, or predicted in the future, based on the weather forecast. This means, the module 2a can plan when to use which energy-procedure to reach the production goals, with least energy consumption and best energy mix and price. The evaluation result comprises a candidate mode of operation 5 that is desirable from the perspective of the module 2a. P240519W001 - 17 - 07.10.2024

[0078] In step d, the module 2a (feeder) communicates the evaluation result to the modules around it, in this example on the other module 2b (reactor).

[0079] After the evaluation for the energy-procedure 3a-3d to be used by themself, the module 2a communicates the result with the downstream modules 2b-2d - the succeeding modules 2b-2d. In the example, the module 2a (feeder) communicates the result to the other module 2b (reactor). Depending on which module 2a, 2b starts with the evaluation, the communication can also be to the upstream modules 2a, 2b.

[0080] In step e, the module 2b (reactor) starts itself the evaluation internally, based on steps a, b and d.

[0081] The modules 2a, 2b have internally another component in the intelligence which receives the evaluation results from the upstream (or downstream) modules 2a, 2b and starts evaluating the energy-procedure 3a-3d for the service to be started depending on the context given by the central control entity 6 and the evaluation result of the upstream modules 2a, 2b. The result comprises whether operation of the chosen module 2* in the candidate mode of operation 5 is possible within the constraints of the other module 2b (OK) or not (NOK).

[0082] In step f, the module 2b (reactor) tells the chosen module 2*, 2a (feeder) if the evaluation made in step c fits with the constraints of itself, i.e. , if it is OK or not (NOK),, from the perspective of the module 2b, that the chosen module 2*, 2a operates in the candidate mode of operation 5.

[0083] If a viable solution is found, the downstream modules further communicate the result to their respective downstream modules, thereby propagating the optimization throughout the industrial plant 1. If no solution is found, this is communicated back to the upstream module 2a, and this restarts the evaluation process with the new constraints given from the downstream modules 2b. If it fits, the module 2a (feeder) can use the procedure. Later the modules 2a-2d switch their procedure synchronously. This procedure is repeated until the best possible energy-procedure that fits the constraints is found.

[0084] Then the created “plan” for when using which procedures 3a-3d is feasible. The lookup table or some Al technique can also be used here in each module 2a-2d to P240519W001 - 18 - 07.10.2024 solve / determine the best solution to avoid too many iterations and communications between feeders 2a and reactors 2b to save more time.

[0085] In step g, the module 2a (feeder) tells the central control entity 6 which energyprocedure 3a-3d of the service is executed as per the desired mode of operation 5, and the energy needed for this.

[0086] Once an optimal “plan” (i.e., an energy plan 1a) for the production is found, the plan is communicated to the POL. The POL can use this to supervise the production and to track the energy efficiency of the process plant. Furthermore, it can communicate the needed energy to a higher-level system to plan the energy production - if present - and to assure that there is always enough energy for the “plan”. In case there is more energy delivered from renewables or for a cheap energy price, the POL can be used to “buffer” this energy in processes where buffering is possible (e.g. produce more process heat and store this heat in heating water or something).

[0087] In step h, during normal operation, the state S of the service is always communicated to the central control entity 6.

[0088] This is the normal modular operation principle. The central control entity 6 knows at any point in time which module 2a-2d is executing which service and the state of the service, as well as the energy-procedure 3a-3d that is executed.

[0089] This procedure is repeated whenever there is a change in the energy information 4 or the services commanded by the central control entity 6. This procedure is propagated through the process between the modules 2a-2d, always between the “neighborhood” modules 2a-2d, meaning the direct predecessor and successor modules 2a-2d.

[0090] The central control entity can communicate the production constraints, e.g. volume per time slice (or minimum and maximum), or final required output at a certain time in the future. This needs to be considered by the modules 2a-2d to reach the production goals / operation targets.

[0091] The function module 2a-2d, 2* to start with in the algorithm is not necessarily the one the is most upstream. The module 2a-2d can be chosen in several different ways: P240519W001 - 19 - 07.10.2024

[0092] 1. The module 2a-2d that the evaluation should start with can be chosen manually.

[0093] 2. The most upstream module 2a can simply be taken as a starting point.

[0094] 3. The module 2a-2d with the highest savings potential is taken as a starting point. Here, the modules 2a-2d start with comparing their savings potential, the one most upstream can start providing its values.

[0095] 4. An algorithm can be used to start iteratively with all modules 2a-2d, get the results and afterwards decide automatically which plan 1a should be taken to have highest saving potential.

[0096] Depending on which module 2a-2d is taken as starting point, the results might differ, since the saving are assumed to be highest for the module 2a-2d that is used as a starting point.

[0097] To sum up, the present method exploits the intelligence of a module that:

[0098] • is aware of the services delivered by the module and the different energyprocedures of every service, including their energy consumptions and outputs,

[0099] • knows how to evaluate a given environmental context in combination with a requested service start for production,

[0100] • knows the process context, meaning the predecessor and successor modules, and

[0101] • can communicate with its process context for negotiation, as well as a communication method that is to:

[0102] • communicate the environmental context from the central control entity 6 to the modules 2a-2f,

[0103] • start / trigger evaluation to adapt the energy-mode,

[0104] • receive acceptance or decline from other modules 2a-2fwith reasoning and unmet constraints (e.g., higher chemical reaction temperature is required, insufficient pressure, longer reaction time, etc.),

[0105] • communicate evaluation results of a module to another module and change mode in sync with other modules, and

[0106] • communicate which energy-procedures are executed in which module back to the central control entity 6. P240519W001 - 20 - 07.10.2024

[0107] The concept presented above allows to dynamically change or adapt the operation targets. In the following, some more examples are given what operation targets could be applied:

[0108] • The production could be shifted to a time where there is a lot of renewable energy provided or where the energy price is predicted to be lower. This would increase sustainability and decrease cost at the same time.

[0109] • The concept works with batch, as well as continuous processes. With batch processes, the target might be a certain amount of production output. With continuous processes, the target might be a throughput per time. The concept works for both.

[0110] • A manufacturing engineering system, MES, could be connected, as well in order to provides further production plans for the future. Then the planning could be even done more efficient. E.g. when the energy price is low, the production could be speeded up and the next production plan could be started earlier in order to produce more during a low-price period.

[0111] • Energy could be stored in processes with buffers. It would be possible to run procedures of modules with a higher energy consumption than needed in order to buffer energy production peeks or “store” energy for later use when the energy price increases, or the energy mix contains less renewable energy.

[0112] It would be possible to first simulate the impact of a mode change of one or several modules before actually doing it. Moreover, the operator may be prompted to approve a mode change of one or more modules. In particular, the “safety mode set” or safety “mode shifting set” may be set beforehand to avoid possible dangerous mode shifting. Also, a warning may be set in that the operator will receive a warning or an approval request when there is dangerous mode shifting. For sure, all dangerous mode shifting should be defined at first.

[0113] In case the discrepancy between the proposal of one module (e.g. run xy throughput) and the capabilities of another module are too high, the operator could be informed. By this, the setup could possibly be adapted to have a more efficient setup from the start next time. P240519W001 - 21 - 07.10.2024

[0114] List of reference signs:

[0115] 1 industrial plant

[0116] 1a energy plan for operating industrial plant 1

[0117] 2a-2d modules or other functional units in industrial plant 1

[0118] 2* chosen module or other functional unit

[0119] 3a-3d modes of operation that differ in energy usage

[0120] 4 energy information

[0121] 5 candidate mode of operation 3a-3d

[0122] 6 central control entity of industrial plant 1

[0123] 100 method for operating industrial plant 1

[0124] 110 obtaining energy information 4

[0125] 120 obtaining candidate mode of operation 5

[0126] 121 choosing module 2a-2f with highest energy consumption

[0127] 122 determining candidate mode of operation 5 by module 2a-2f

[0128] 123 communicating production constraint from central control entity 6

[0129] 124 determining candidate mode of operation 5 given production constraint

[0130] 130 determining whether candidate mode of operation 5 is usable

[0131] 131 negotiating use of candidate mode of operation 5 between modules 2a-2f 131a communicating constraint back to chosen module 2*

[0132] 131b determining new candidate mode of operation 5 given constraint

[0133] 132 checking ability to deliver required contribution to industrial process

[0134] 132a checking sufficient stock of material

[0135] 133 checking ability to deliver output

[0136] 133a checking buffer capacity

[0137] 134 checking whether production goal of industrial process can be met

[0138] 134a checking by last module 2a-2f producing product of the process

[0139] 140 authorizing candidate mode of operation 5 in energy plan 1a

[0140] 141 operating modules 2a-2f in operating modes determined during engineering

[0141] 142 determining energy plan 1a for predetermined period of time in the future

[0142] 150 branching back to step 120 with new chosen module / functional unit 2*

[0143] 160 determining whether criterion for finishing optimization fulfilled

[0144] 161 using as criterion that all modules 2a-2f have been considered

[0145] 162 using as criterion that energy / cost savings goal has been met P240519W001 - 22 - 07.10.2024

[0146] 170 applying energy plan 1a to industrial plant 1

[0147] C command to start services

[0148] E educt of industrial process

[0149] P product of industrial process S service states of modules 2a-2f

[0150] SP currently executed service procedure of modules 2a-2f

Claims

P240519W001 - 23 - 07.10.2024Claims:

1. A method (100) for operating an industrial plant (1) that is configured to execute an industrial process using a plurality of interconnected process modules or other functional units (2a-2f), wherein at least one process module or other functional unit (2a-2f) is capable of performing its assigned role in the industrial process in different modes of operation (3a-3d) that have different energy consumptions, and wherein the method (100) comprises the steps of:• obtaining (110) energy information (4) that is indicative of the availability, and / or of the cost, of energy required for the operation of at least one process module or other functional unit (2a-2f);• obtaining (120), for at least one chosen process module or other functional unit (2*, 2a-2f), a candidate mode of operation (5) that is desirable from the point of view of this process module or other functional unit (2*, 2a-2f) with respect to energy consumption;• determining (130) whether operation of the chosen process module or other functional unit (2a-2f) if this candidate mode of operation (5) is compatible with the intended operation of one or more further process modules or other functional units (2a-2f) that are connected to the chosen process module (2*, 2a-2f); and• if this is the case, authorizing (140), in an energy plan (1a) for operating the industrial plant (1), operation of the chosen process module or other functional unit (2*, 2a-2f) in the candidate mode of operation (5).

2. The method (100) of claim 1, further comprising: branching back (150) to obtaining a candidate mode of operation (5) for a next chosen process module or other functional unit (2*, 2a-2f).

3. The method (100) of any one of claims 1 to 2, wherein, for each process module or other functional unit (2a-2f) for which operation in a candidate mode of operation (5) has not been authorized, the energy plan (1a) for operating the industrial plant comprises (141) operation of this process module or other functional unit (2a-2f) in a predetermined mode according to the engineering of the industrial plant (1).

4. The method (100) of any one of claims 1 to 3, further comprising: in response to determining (160) that a predetermined criterion for the finishing of the energyP240519W001 - 24 - 07.10.2024 plan (1a) is met, causing (170) operation of the industrial plant (1) according to the energy plan (1a).

5. The method (100) of claim 4, wherein the predetermined criterion for the finishing of the energy plan comprises that:• all process modules or other functional units (2a-2f) of the industrial plant (1) have been chosen process module or functional unit (2*) in the course of the method (161); and / or• operating of the industrial plant (1) according to the energy plan (1a) causes a predetermined goal in terms of energy and / or cost savings to be met (162).

6. The method (100) of any one of claims 1 to 5, wherein a process module or functional unit (2a-2f) with a highest energy consumption is chosen (121) as the chosen process module or functional unit (2*).

7. The method (100) of any one of claims 1 to 6, wherein the desirable candidate mode of operation (5) is determined (122) by the chosen process module or functional unit (2*, 2a-2f).

8. The method (100) of claim 7, wherein the determining whether operation of the chosen process module or other functional unit (2*, 2a-2f) in the candidate mode of operation (5) is compatible with the intended operation of one or more further process modules or other functional units (2a-2f) that are connected to the chosen process module (2*, 2a-2f) comprises (131) a negotiation between the chosen process module or other functional unit (2*, 2a-2f) and the further process modules or other functional units (2a-2f).

9. The method (100) of claim 8, wherein the negotiation comprises:• communicating (131a), from the further process module or other functional unit (2a-2f), to the chosen process module or other functional unit (2*, 2a- 2f), at least one constraint by virtue of which operation of the chosen process module or other functional unit (2*, 2a-2f) in the candidate mode of operation (5) is not compatible with the intended operation of the further process module or other functional unit (2a-2f), and• determining (131b), by the chosen process module or other functional unit (2*, 2a-2f), a new candidate mode of operation (5) given the at least one constraint.P240519W001 - 25 - 07.10.202410. The method (100) of any one of claims 7 to 9, wherein energy information (4) is provided to the chosen process module or other functional unit (2*, 2a-2f) by a central control entity (6) of the industrial plant (1).

11. The method (100) of any one of claims 7 to 10, wherein• a central control entity (6) of the industrial plant (1) communicates (123) at least one production constraint to the chosen process module or functional unit (2*, 2a-2f); and• the chosen process module or functional unit (2*, 2a-2f) determines (124) the desirable candidate mode of operation (5) given this production constraint.

12. The method (100) of any one of claims 1 to 11, wherein the determining whether operation of the chosen process module or other functional unit (2a*, 2a-2f) in this candidate mode of operation (5) is compatible with the intended operation of one or more further process modules or other functional units (2a- 2f) that are connected to the chosen process module (2*, 2a-2f) is made dependent on one or more of:• whether a further process module or other functional unit (2a-2f) that is connected downstream of the chosen process module or other functional unit (2a-2f) is able (132) to deliver its required contribution to the industrial process despite a reduced output rate of the chosen process module or other functional unit (2*, 2a-2f);• whether a further process module or other functional unit (2a-2f) that is connected upstream of the chosen process module or other functional unit (2*, 2a-2f) is able (133) to deliver its output despite a reduced intake rate of this output by the chosen process module or other functional unit (2*, 2a-2f); and• whether a goal of the industrial process as a whole can still be met (134) despite operation of the chosen process module or other functional unit (2*, 2a-2f) in the candidate mode of operation (5).

13. The method (100) of claim 12, wherein a process module or other functional unit (2a-2f) that delivers a final product of the industrial process determines (134a) whether the goal of the industrial process as a whole can still be met.

14. The method (100) of any one of claims 12 or 13, wherein• whether a further process module or other functional unit (2a-2f) that is connected downstream of the chosen process module or other functionalP240519W001 - 26 - 07.10.2024 unit (2*, 2a-2f) is able to deliver its required contribution to the industrial process despite a reduced output rate of the chosen process module or other functional unit (2*, 2a-2f) is made dependent on whether the further process module or other functional unit has (132a) a sufficient stock of the output of the chosen process module or other functional unit (2*, 2a-2f), and / or• whether a further process module or other functional unit (2a-2f) that is connected upstream of the chosen process module or other functional unit (2*, 2a-2f) is able to deliver its output despite a reduced intake rate of this output by the chosen process module or other functional unit (2*, 2a-2f) is made dependent on whether there is (133a) sufficient buffer capacity for accepting the output of the further process module or other functional unit (2*, 2a-2f).

15. The method (100) of any one of claims 1 to 14, wherein the energy plan (1a) is determined (142) for a predetermined period of time in the future.

16. The method (100) of any one of claims 1 to 15, wherein• multiple candidate energy plans (1a) are determined (143) starting from different chosen process modules or other functional units (2*, 2a-2f), and• a candidate energy plan (1a) that is optimal with respect to a given optimization goal is chosen (144) as the final energy plan (1a).

17. A computer program, comprising machine-readable instructions that, when executed by one or more computers and / or compute instances, cause the one or more computers and / or compute instances to perform the method (100) of any one of claims 1 to 16.

18. A non-transitory machine-readable data carrier, and / or a download product, with the computer program of claim 17.

19. One or more computers and / or compute instances with the computer program of claim 17, and / or with the machine-readable data carrier and / or download product of claim 18.

Images