Method, system and apparatus for controlling compressor system

Abstract

Methods, systems, and apparatus are provided for improving control and efficiency of a compressor system. The compressor system includes a controller. The controller is configured to predict a future demand for the compressor system, determine an initial sequence based on the predicted future demand, determine an initial set switching time for the initial sequence, optimize the initial switching sequence based on the set switching time to form an optimized switching sequence, and output the optimized switching sequence to the compressor system. Iteratively determining a set of switching times for the optimized switching sequence and optimizing the optimized switching sequence based on the set of switching times until a final switching sequence and a final set of switching times are obtained, and controlling operation of the set of components based on the final switching sequence and the final set of switching times.

Description

Technical field

[0001] Disclosed herein are methods, systems, and apparatus for monitoring and controlling a compressor system, and in particular, for monitoring, controlling, and optimizing the efficiency of components of a compressor system that provides compressed air or gas to a consumer. [Background Technology]

[0002] Compressors are known for compressing air or gas in one or more compression stages. The compressed air or gas is then supplied to one or more consumers. Distribution of the compressed air or gas can be provided via a compressed air or gas system.

[0003] Since the number of consumers may be large and spatially dispersed in important areas such as in a factory or a hospital, a central hub is usually set up to provide the compressed air or gas.

[0004] A central hub typically includes one or more compressor rooms, each housing one or more compressors. Auxiliary equipment such as valves, filters, dryers, containers, sensors, control components, and / or other devices used to manage and / or control the compressor rooms are also installed. Pipelines or conduits then extend from the one or more compressor rooms to supply the consumers. As the final step in the chain, the compressed air or gas is then distributed to various applications by the consumers.

[0005] Furthermore, between the compressors and the consumers, there may also be another set of devices, such as safety valves, distribution valves, control sensors, or other devices configured to control and protect the distribution of the compressed air or gas.

[0006] The mechanism will further be identified as a compressor system. Thus, a compressor system may include a compressor that supplies a consumer, but is generally considered to be broader, such that it includes numerous components and constitutes a complex system in which several components interact with each other.

[0007] In order to utilize the compressor system, its different components need to be controlled. It is known to control the compressors separately by means of independent zone controllers, whereby the different controllers are set to predetermined pressure values, thereby sequentially turning the compressors on or off depending on the compressed air consumption.

[0008] It is also known to employ a method of controlling a compressor system using multiple communicating controllers configured to control multiple components that may be part of the compressor system, thereby controlling the components such that no single controller determines the operating conditions of any component controlled by another controller. Such a method is disclosed in International Patent Application No. WO2008 / 009073.

[0009] In the international patent application WO 2008 / 009072, another method for controlling a compressed air unit is disclosed, which is composed of several compressed air or gas networks, which have at least one commonly controllable component, whereby at least this common component is controlled by at least one controller based on measurement data of at least one of the compressed air or gas networks.

[0010] However, a disadvantage of these control methods identified by the inventors is that they operate purely on the basis of the current state of the compressor system, meaning that they cannot take into account any type of forecasting. This leads to suboptimal control and higher energy costs.

[0011] The compressor system can be presented as a switched dynamic system, a continuous-time nonlinear system defined by multiple subsystems and nonlinear switching rules. Switched dynamic systems show great flexibility in modeling a wide range of real-world applications, but due to the discrete nature of the switching dynamics, it has proven difficult to implement beneficial control of such systems.

[0012] It has been identified by the inventors that the problem in controlling a switching dynamic system can take the following form:

[0013] where x(t) is the state of the system, and u(t) and v(t) are the inputs. The switching nature of these systems is captured in the variable u(t), which can take values ​​from only a finite, discrete set of options. This is known as the mixed-integer optimal control problem (MIOCP).

[0014] Known approaches for addressing MIOCP include mixed-integer nonlinear programming (MINLP), relaxation solutions, control parameterization techniques (CPET), and combined integral approximations (CIA). However, the inventors have identified that currently known approaches exhibit significant drawbacks. Specifically, the inventors have identified that such approaches and systems are overly complex, require significant computational capacity and latency due to the required calculations, and may be inaccurate.

[0015] It has been discovered by the inventors of the present application that known calculations and optimizations are indeed unnecessary and inefficient, requiring more time and processing capacity than necessary, and the inventors of the present application have discovered a robust and efficient method and system to accurately control and optimize the performance efficiency of a compressor system. [Summary of the invention]

[0016] A compressor system is provided, comprising: a group of components fluidly connected to a shared compressed air distribution network; and a controller configured to: predict future demand for the compressor system; determine an initial switching sequence for the operation of at least one component of the group of components that meets the predicted future demand; determine a set of switching times for the initial switching sequence; optimize the initial switching sequence based on the set of switching times to form an optimized switching sequence; iteratively determine a set of switching times for the optimized switching sequence, and optimize the optimized switching sequence based on the set of switching times until a final switching sequence and a final group switching time are obtained; and control the operation of the group of components based on the final switching sequence and the final group switching time.

[0017] In another embodiment, a controller for a compressor system is provided. The controller is configured to operate a compressor system having a set of components fluidly connected to a shared compressed air distribution network, the controller comprising: a computer-readable storage medium; and a processor configured to: predict a future demand for the compressor system; determine an initial switching sequence for operation of at least one component of the set of components that meets the predicted future demand; determine a set of switching times for the initial switching sequence; optimize the initial switching sequence based on the set of switching times to form an optimized switching sequence; iteratively determine a set of switching times for the optimized switching sequence and optimize the optimized switching sequence based on the set of switching times until a final switching sequence and a final set of switching times are obtained; and control the operation of the set of components based on the final switching sequence and the final set of switching times.

[0018] In another embodiment, a computer-implemented method for controlling a compressor system to improve the efficiency of the compressor system is also provided, the compressor system comprising a set of components fluidly connected to a shared compressed air distribution network. The method comprises: predicting a future demand for the compressor system; determining an initial switching sequence for operation of at least one component of the set of components that meets the predicted future demand; determining a set of switching times for the initial switching sequence; optimizing the initial switching sequence based on the set of switching times to form an optimized switching sequence; iteratively determining a set of switching times for the optimized switching sequence and optimizing the optimized switching sequence based on the set of switching times until a final switching sequence and a final set of switching times are obtained; and controlling the operation of the set of components based on the final switching sequence and the final set of switching times.

[0019] A hardware storage device having stored thereon a plurality of computer-executable instructions, which, when executed by one or more processors of a computing system, configure the computing system to implement a method for controlling a compressor system, the method comprising predicting a future demand for the compressor system; determining an initial switching sequence for the operation of at least one component of the group of components that meets the predicted future demand; determining a set of switching times for the initial switching sequence; optimizing the initial switching sequence based on the set of switching times to form an optimized switching sequence; iteratively determining a set of switching times for the optimized switching sequence, and optimizing the optimized switching sequence based on the set of switching times until a final switching sequence and a final group switching time are obtained; and controlling the operation of the group of components based on the final switching sequence and the final group switching time.

Brief Description of the Drawings

[0020] Figure 1 A specific embodiment of a compressor system is shown.

[0021] Figure 2 Shown from Figure 1 A specific embodiment of a controller of a specific embodiment. Figure 3 Shown from Figure 2 A specific embodiment and further details of model predictive control of a specific embodiment.

[0022] Figure 4 Shown from Figure 3 Prediction of a specific embodiment A specific embodiment and further details in the future.

[0023] Figure 5 An embodiment of a method for iterative switching time optimization is shown. The figures are included to provide a better understanding of the components and are not intended to be limiting in scope but rather to provide exemplary illustrations. [Specific implementation method]

[0024] The inventive concepts disclosed herein are described below with reference to specific embodiments and accompanying drawings. However, the scope of the claims is not limited thereto. The drawings are merely schematic and non-limiting. In the drawings, the size of some components may be exaggerated and not drawn to scale; this is for ease of illustration. Dimensions and relative sizes do not necessarily correspond to practical embodiments of the present invention.

[0025] Furthermore, the terms first, second, third and the like may be used to distinguish similar elements and do not necessarily describe a sequential or chronological order. These terms are interchangeable under appropriate circumstances, and embodiments of the invention may be practiced in an order other than described or illustrated herein.

[0026] In addition, various embodiments that may be described as "preferred embodiments" are to be construed as merely illustrative of the manner and mode for carrying out the invention and are not to be construed as limiting the scope of the invention.

[0027] The terms "comprises," "includes," or "having" used in the claims should not be construed as limiting the means or steps mentioned thereafter. These terms are to be interpreted as specifying the presence of the indicated features, components, steps, or elements, but do not preclude the presence or addition of one or more other features, components, steps, or elements, or groups thereof. Thus, the scope of the expression "a device or apparatus includes means A and B" should not be construed as limited to "a device or apparatus consisting solely of components A and B." For the purposes of this disclosure, specific reference to components A and B of the apparatus is intended, but the claims should be further inferred to encompass equivalents of such components.

[0028] Generally speaking, the compressor systems disclosed herein include one or more compressors configured to provide compressed air or gas to a customer network. As described herein, a compressor is provided as a compressed gas source, but the compressor system may include other compressed gas sources, such as a pre-compressed gas tank, a reservoir, or a supply line or pipeline. The compressor systems may further include containers or tanks for storing compressed air or gas, and valves connected to the customer network, which may have one or more consumers. Additional devices may also be included, such as dryers, filters, regulators, and / or lubricators.

[0029] Figure 1 The figure shows a compressor system 100 including three compressors 101, 101' and 101" configured to provide compressed air or gas to a customer network 105. The compressor system 100 further includes a container or tank 103 for storing compressed air or gas, and a valve 104 connected to the customer network 105. At the customer network 105, there is one or more consumers. It should be further understood that the compressor system 100 may further include other devices, such as dryers, filters, regulators, and / or lubricators, as indicated above, but in the structure of this document, reference will be made to the compressor system 100. Figure 1 The specific embodiment is shown as a compressor system 100. Figure 1 , solid lines indicate fluid connections, while dashed lines indicate data connections.

[0030] The compressors 101, 101', 101" can each be regionally controlled by a separate controller 102, 102', 102". In addition, in order to efficiently control the compressor system 100, the controllers 102, 102', 102" can be controlled in a coordinated manner. In other words, it is possible to avoid each controller 102, 102', 102" individually controlling its respective compressor 101, 101', 101". However, the controllers 102, 102', 102" can be commanded by a controller 106, so that the overall performance and efficiency of the compressor system 100 are increased.

[0031] The controllers 102, 102', 102", 106 may include a processor, such as a microprocessor, a memory device, an output interface, and an input interface. The controllers 102, 102', 102", 106 may be configured to receive input signals, which may be received via wired or wireless means, through the input interface and process received sensor signals obtained from components and associated sensors within the compressor system 100. And as described herein, the controllers 102, 102', 102", 106 output control signals to the components of the compressor system 100 through the output interface. As described in more detail below, based on the iterative STO determination of the optimal schedule for the compressor system 100, the control signal is transmitted at the controller 106 to adjust the operating parameters of the compressor system. In certain specific embodiments, the controller 100 is configured to transmit control signals to the controllers 102, 102', 102" to adapt the operation of the compressors 101, 101', 101", or to turn the compressors 101, 101', 101" on or off depending on the optimal schedule determined by the controller 106.

[0032] The controller 106 may be located near the controllers 102, 102', 102", but may also be located at a remote location compared to the other components of the compressor system 100. For example, the controller 106 is not necessarily integrally formed with or connected to the compressor system 100. The controller 106 may be located in proximity to the compressor system 100, such as within the same room volume, or within a house. Alternatively, the controller 106 may be remote from the compressor system and its components while still being able to receive signals from and transmit signals to the components of the compressor system 100. In addition, the controller 106 may be communicatively connected to a remote computer system for purposes such as remote monitoring, control, adjustment and / or software updates, and data obtained by the controller or control unit 106 and operating parameters transmitted by the controller or control unit 106 as control signals may be transmitted to the remote computer system or data storage device for further analysis and / or processing.

[0033] The controller or control unit 106 may include or utilize a dedicated or general purpose computer system, or a computing unit that includes computer hardware, such as a processor or one or more processors and system memory, specifically within the control unit or controller 106 or alternatively in communication with the controller 106, as discussed in greater detail below. The controller 106 may be located relatively close to the compressor system 100 and receive hardwired or wireless signals from other components of the compressor system 100 and transmit hardwired or wireless signals to other components of the compressor system 100. Alternatively, the controller 106 may be located remotely from the other components of the compressor system and receive signals from the other components of the compressor system, including from one or more sensors that can provide an indication of one or more operating characteristics in the compressor system, and transmit the signals to the other components of the compressor system over a network, such as a local area network (LAN), a wide area network (WAN), the Internet, or some other network. Alternatively, one of the controllers 102 , 102 ′, 102 ″ may be configured to function as the controller 106 for controlling all of the compressors 100 , 100 ′, 100 ″.

[0034] Through the controller 106, the operation, switching, and idle costs of the compressor system 100 can be managed, thereby reducing wear on components of various devices and simultaneously reducing or otherwise improving the energy consumption of the compressor system 100. To this end, the controller 106 can be configured to optimally schedule the operation of the components of the compressor system 100 according to various embodiments disclosed herein.

[0035] In the disclosed concept, controller 106 receives characterization data 110 describing technical or functional properties of one or more components of compressor system 100. This characterization data may be obtained from a database, a model, measurements on one or more components configured for compressor system 100, or any other suitable means. Controller 106 also receives prediction data 120 describing at least future predicted airflow and / or pressure demand of client network 105. Again, this prediction data may be obtained from a database, a model, measurements on one or more components configured for compressor system 100 or client network 105, or any other suitable means. Based on characterization data 110, prediction data 120, and an action profile determined for compressor system 100 according to the present disclosure, controller 106 may transmit configuration data 130 to controllers 102, 102', 102" to coordinate control of compressors 101, 101', 101".

[0036] An example of how the controller 106 may control the compressor system 100 is shown in FIG. Figure 2In the depicted embodiment, the controller 106 controls and communicates with the compressor system 100 via an output 210 and, optionally, an input 211. Various modules or components of the controller 106 may be configured to provide data to a model predictive control (MPC) block 205 to determine the output 210.

[0037] According to various embodiments, the controller 106 may include a database 200, a set of compressor models and / or compressor system models 201, one or more estimators 202, a flow prediction block 203, and a sampling block 204 for providing an initial sequence to an MPC block 205. Although these blocks 200, 201, 202, 203, 204, and 205 are shown as part of the controller 106, it should be noted that these blocks can be physically or even virtually distributed relative to each other. For example, the database 200 can be located on a remote server and accessed via a customized data connection. Similarly, it should be noted that the controller 106 can include more or fewer than these blocks 200, 201, 202, 203, 204, and 205 and can be configured in various architectures for determining the output 210.

[0038] like Figure 2 As shown in FIG. 1 , one or more estimators 202 may be configured to receive 220 measurements 211 of the compressor system 100. One or more estimators 202 may receive additional input 221 from the database 200 and may, for example, use an existing set of compressor models 201 as another input 222. In various embodiments, the set of compressor models 201 may also be incorporated 223 into the database 200 itself. The set of compressor models 201 may be representative of the compressor system 100. For example, the model may be a digital twin of the compressor system, may include a set of differential equations representing the compressor system, or may include a black box approach.

[0039] The estimator block 202 can estimate the current state of the compressor system 100 based on the measurements 211 received 220 and, optionally, based on the model 201. Furthermore, previous estimates 221 can be uploaded from the database 200 to increase the accuracy of the estimate. The output of the estimator block 202 can be used as input 224, 227 to the flow prediction block 203 and / or the MPC block 205.

[0040] The prediction block 203 can be configured to predict one or more future process variables of the compressor system 100. The prediction 225 can be based on the output 224 of the estimator block 202 and on data 226 stored in the database 200. The prediction block 203 can use the current process variables of the compressor system 100 and the agent state data to calculate the expected state of the compressor system 100 for an appropriate timeframe. For example, tank pressure and flow demand can be expressed as predicted future demands on the compressor system 100 in the future process variable profile.

[0041] According to the present disclosure, the terms "forecast," "predict," "predicted," and similar terms refer to estimating outcomes for unknown data. Forecasting is a subfield of prediction that uses time series data to make predictions about the future. For example, the difference between forecasting and prediction is that the latter considers the time dimension. Therefore, the term "forecast" can also be interpreted as "forecast," and the term "forecast" will be used throughout this disclosure.

[0042] The prediction block 203 may be configured to consider past process variable data via 226 the database 200 and current process variable data via 224 the estimator block 202. Additionally, other input data may be considered, such as sensor data, past and future state proxy data, production schedules, calendar data, holiday data, and / or weather forecast data.

[0043] The output 225 of the prediction block 203 comprises a data profile of the process variables for a given prediction at a predetermined timeframe which may be set by the user or by an MPC program as will be discussed further. In the MPC program, setting of the timeframe may be automated.

[0044] The prediction block 203 may be a predictor function block based on an input-output model having inputs, outputs, model parameters, and hyperparameters. As illustrative examples, four prediction paradigms suitable for the prediction block 203 are discussed.

[0045] First, a multiple output prediction strategy can be used, which uses any function approximator to directly estimate or train the predictor function for a given fixed time horizon H. This approach is also called a multi-step approach, in which the multivariate predictor function is directly trained given current and past observations. Second, a recursive multi-step prediction method can be used, in which a suitable (I) / O model is selected. From the training parameters of the (I) / O model, a predictor function is constructed, and the output can be recursively simulated or forecasted for a given time horizon H. Third, a direct multi-step prediction strategy can be used, which includes a separate predictor structure for each forecast period. As a fourth prediction paradigm, a hybrid prediction strategy combining two or more of the above-mentioned paradigms can be used. Of course, other prediction strategies that are clear to those skilled in the art from the present disclosure can be used.

[0046] In sampling block 204, output 225 may be sampled at a sampling frequency suitable for MPC block 205. The sampling frequency may be reset or varied in time if necessary.

[0047] Figure 3 FIG. 1 shows a flow chart of an MPC block 205 configured to control the compressor system 100 according to various embodiments disclosed herein. Figure 3 As shown in FIG, the MPC block 205 operates based on one or more demands 300 , one or more constraints 301 , and a future forecast 302 of demand for the compressor system 100 .

[0048] Requirements 300 may include, for example, a constant pressure or a constant flow rate in client network 105. Constraints may include, for example, demand constraints such as pressure limits or setpoint pressures, flow requirements for a mixture or a portion of a mixture, humidity limits, temperature limits, particle limits, or limits on other impurities such as oil in the output fluid, dissolved oxygen in the process, or the like. Constraints may further include, for example, system constraints such as maximum and / or minimum temperature limits, humidity limits, flow limits, impurity limits, speed limits, acceleration limits, jerk limits, valve limits and rates of change of valve positions, vibration limits, current limits, sequencing between elements of the system, and time constraints such as between starts, between stops, minimum time in a state, maximum time in a state, delayed second stops, or the like.

[0049] These system constraints may be included at any location or component of the system. For example, system constraints may include maximum and minimum temperature limits at the inlet of the air utility or supercharger, at motor components such as windings or converters, at compressor components, at cooling system water, at compressor oil, at the outlet of the compressor for the energy recovery system, or the like. Humidity limits, flow limits, and impurity limits may be included at the inlet of the air utility, supercharger, and the like, for example.

[0050] Furthermore, any of the important set points and / or the mentioned constraints can be tracked to improve the quality of the output air and, as a result, the quality of the end product. In a related concept, weighting can be used to create combinations of the mentioned constraints in the same framework without requiring significant adaptation.

[0051] In view of the identified constraints, embodiments of the compressor system may include one or more sensors located at a plurality of predetermined locations within the system. For example, the compressor system may include any one or a combination of a temperature sensor, a humidity sensor, a flow sensor, a speed sensor, an acceleration sensor, an image sensor, a current sensor, a vibration sensor, a particle sensor, an oxygen sensor, a nitrogen sensor, a position sensor, a pressure sensor, a pressure dew point sensor, a rotational speed sensor, and related components.

[0052] As will be understood by those skilled in the art from the present description, the compressor system can be configured to include any known sensors related to the compressor system and / or compressed gas. For example, the temperature sensor can include one or more thermocouples, liquid or gas thermometers, electrical thermometers including, for example, resistance thermometers, silicone diodes, bimetallic devices, bulb and capillary sensors, sealed bellows, and / or radiation thermometry devices, or any other type of temperature sensing device. Furthermore, one or more sensors of the compressor system can be remotely located from a wall or sidewall of a component of the system, such as a pressure vessel, or a pipeline or conduit, while still obtaining their respective sensor data based on, for example, radiation thermometry or other remote sensing means.

[0053] Any of the aforementioned sensors may be provided with a means for communicating with controller 106. The communication connection may be wireless or wired; for clarity, the sensors and associated communication sensors are not shown. Individual output signals or data from these sensors are transmitted to controller 106 via hardwiring or wireless communication and may be further used by controller 106 to adjust or modify inputs to determine an optimal schedule for the compressor system and / or track operational characteristics of the system. The described embodiments further or alternatively include writing sensor and / or constraint data to memory. The memory may be a component of controller 106 or a component external to the controller.

[0054] In the presently disclosed embodiment, the MPC block 205 can employ iterative switching time optimization (STO) to define an optimal sequence and switching timing for the compressor system in the form of an action profile 320 or schedule for the compressor system 100. The action profile 320 can include a plurality of instructions for improved operation of the compressor system 100 such that the energy demand of the compressor system 100 and wear on the components of the compressor system 100 are reduced. The iterative STO can include an STO module 310 and an optimization module 311 for determining the action profile 320, as discussed in more detail below.

[0055] According to the present disclosure, the controller is configured to determine an action profile for the compressor system using an MPC framework that takes into account future customer demand. Figure 4 As shown in the diagram of FIG, setpoints 405, past data 410, models, and predicted demands may be used to form inputs to the MPC. Past data 410 may include past setpoints 402 and their actual values ​​403, as well as actions previously taken by the compressor system 404. In various configurations, the models may include static or dynamic machine models, static or dynamic airnet models, and / or instantaneous flow demands versus periodic flow demands. Predictions 406 may be generated for parameters under control, as well as a restricted subset of the compressor system's conditions, such as compressor or other component conditions, resulting flows and pressures at time periods k to k+n for a prediction period 409, predictions of inlet air or atmospheric conditions based on weather information, or the like. These inputs may be provided as part of an initial sequence of the presently disclosed iterative STO, or may be provided with the initial sequence, and may be prepared, for example, using dynamic programming, analytical dynamic programming (ADP), artificial intelligence (AI), heuristic, a branch and bound scheme, a linear program simplex solver, or the like.

[0056] Using this initial sequence 407, STO can be applied according to Figure 5The method of determining the optimal switching time for the compressor system's operating profile is used. The initial sequence is provided to the STO module 310 in the first step 501 of method 500. As previously explained, the initial sequence can be provided using dynamic linear programming, ADP, AI, heuristics, branch-and-bound, linear programming single solver, or similar methods. In the second step 502, the STO module establishes the switching time as a variable to be optimized for the compressor system based on the assumptions of the initial sequence. In the third step 503, the STO module optimizes the cost including the constraints and requirements of the compressor system to determine the time value for each part of the initial sequence 407.

[0057] The STO module calculates a set of switching times based on the following equations, constraints, and requirements:

[0058]

[0059] Subj.

[0060]

[0061] p low ≤p(t)≤p high

[0062] Q low ≤Q k (t)≤Q high

[0063] t low ≤t(S on，k , S lo，k )≤t high

[0064] S on，k , S lo，k ∈{0,1}

[0065] In this example, c k Represents the capacity of a unit, S on,k Represents the operating state, S lo,k Represents the load state, P k represents the power that varies with the capacity and state of unit k, p(t) represents the pressure of the system, represents the first derivative of the pressure, f represents the system dynamics, t(S on,k ,S lo,k ) represents the timing constraint, t0 represents the start time (now), t f Represents the final time or end time of the period.

[0066] According to various embodiments, the STO module can calculate a set of switching times based on the following equations, constraints, and requirements, at least in part using similar notations as described above for constants, variables, and parameters:

[0067]

[0068] p low ≤p(t)≤p high

[0069]

[0070] Q low ≤Q k (t)≤Q high

[0071] t low ≤t(S on，k , S lo，k )≤t high

[0072] S on，k , S lo，k ∈{0,1}

[0073] Further alternatives, variations and combinations are of course contemplated and not excluded from the present disclosure.

[0074] Based on the STO results, the optimization module 311 identifies any portions of the initial sequence 407 where the STO indicates that a new sequence should be formed by removing these identified portions from the initial sequence 407 in an optimization step 504. In another step 505 of method 500, the optimization module communicates the new sequence to the STO module 310 for another iteration of warm starting based on the results of the previous iteration. The method can be iteratively implemented until an optimal sequence and optimal switching times are determined, which form an action profile or schedule for the operation of the compressor system.

[0075] Advantageously, the iterative STO of the described embodiments allows for more accurate start and stop timing in the compressor system than known methods, allows for inclusion of a larger set of constraints, and can handle a wider range of air utilities and targets.

[0076] The iterative approach disclosed herein recognizes two subproblems for optimizing the schedule of a compressor system: (1) finding a switching sequence and (2) optimizing the switching instances given a switching sequence. Advantageously, any remaining problem can be formulated as a continuous linear program for which efficient numerical algorithms exist.

[0077] In the embodiment disclosed in this case, a sequence can be defined as a set of ordered states S={s0,s1,...,s ns In order to specify the switching instance, the set W = {w0, w1, ..., w ns} can be based on the system in state s i w is defined by the time spent in i The reason is that STO determines W under a given S.

[0078] Surprisingly, in order to find the optimal sequence according to the described embodiment, the method starts with an initial sequence S that is not claimed to be optimal and is not based on a relaxation solution. Instead, the method advantageously only needs to include the optimal sequence as a set. Additional states that are not part of the optimal sequence are iteratively removed from this initial sequence by the STO and optimization of the described method. In other words, given a sequence, STO will be used to find the optimal switching times, and given these optimal switching times, the sequence can then be optimized. This is done iteratively: after each removal, STO is solved again for the new sequence, thereby identifying further candidates for removal. Since the initial sequence is finite, the iteration ends when the optimal sequence and the optimal switching time are identified.

[0079] While not being bound by any particular theory, it is believed that the efficiency of the presently disclosed iterative STO method stems from its transformation of the mixed-integer problem into a continuous one, allowing numerous constraints to be introduced naturally and simply, such as w>0.5 to impose a minimum available time constraint on the compressors of the compressor system. Because the maximum number of switches is already limited by the initial sequence, there is no risk of frequent switching. It is also possible to impose conditional constraints, such as not setting u1=1 unless u2=1 or u3=1 is previously set, because these conditional constraints can be directly applied to the sequence to filter out prohibited combinations.

[0080] The method of the present application may initially appear disadvantageous due to the large number of possible combinations that may be required for the initial sequence. However, surprisingly, the inventors have discovered that this concern is unfounded. In particular, the initial sequence only needs to contain the optimal sequence, and the power set grows exponentially. Furthermore, an increase in the number of variables is not necessary. For example, in the multi-shot method, STO fixes u(t) and replaces it with a small amount w, which reduces the number of variables compared to the prior art relaxation problem.

[0081] Specific embodiments of the sequence optimization method can be configured to iteratively select only one state as a candidate for removal, or to perform parallel processing of multiple states to reduce the number of required iterations. Similarly, a branch-and-bound approach can be implemented for selecting between constraint satisfaction and state removal to ensure sequence optimality. Finally, sequence optimization can be configured to insert necessary states rather than remove them. As will be readily apparent from the advantages described above, the methods and systems of the present disclosure not only reduce wear, energy requirements, and system runtime on the compressor system, but also provide greater flexibility than those known in the prior art.

[0082] The special approach to handling the scheduling of the compressor system's components according to the present embodiment can result in significant improvements over the state of the art. The described iterative STO can advantageously handle a wider range of actuating units or components than conventional approaches, including compressors, such as positive displacement compressors, turbo compressors, superchargers, blowers (low pressure), and the like; air utilities, such as dryers, valves, aftercoolers, chillers, oxygen generators, nitrogen generators, and the like; cooling circuits or oil cooling circuits; and energy recovery systems. The described iterative STO can further account for passive air utility components, such as filters, containers, piping, etc., which is not possible with known methods and systems.

[0083] The embodiments further provide the additional advantages of iterative STO, which include the ability to create multiple predictions of, for example, humidity, temperature, impurities, and the like, and the ability to accurately process long time periods. For example, the prediction period according to the present disclosure can be up to 6 hours, up to 8 hours, up to 10 hours, and preferably up to 8 hours. Similarly, the accuracy of iterative STO of the methods and systems can be between 0.5 seconds and 5 minutes, more specifically between 0.5 seconds and 3 minutes, or between 3 seconds and 3 minutes, between 10 seconds and 2.5 minutes, less than 5 minutes, less than 4 minutes, less than 3 minutes, less than 2 minutes, less than 1 minute, less than 45 seconds, less than 30 seconds, less than 10 seconds, or less than 5 seconds.

[0084] Embodiments within the scope of the present disclosure also include physical and other computer-readable media for carrying or storing computer-executable instructions and / or data structures. Such computer-readable media can be any available media that can be accessed by a general-purpose or special-purpose computer system. Computer-readable media that store computer-executable instructions and / or data structures is computer storage media. Computer-readable media that carry computer-executable instructions and / or data structures is transmission media. Thus, by way of example, embodiments disclosed herein may include at least two distinct types of computer-readable media: computer storage media and transmission media.

[0085] Computer storage media is a physical storage medium that stores computer-executable instructions and / or data structures. Physical storage media includes computer hardware, such as random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), a solid-state drive ("SSD"), flash memory, phase-change memory ("PCM"), optical disk storage, magnetic disk storage or other magnetic storage devices, or any other hardware storage device(s) that can be used to store program code in the form of computer-executable instructions or data structures, which may be contained in the controller 106, a general-purpose or special-purpose computer system, or accessed and executed by the controller, the general-purpose or special-purpose computer system to implement the disclosed functionality of the present disclosure.

[0086] Transmission media can include networks and / or multiple data links that can be used to carry program code in the form of computer-executable instructions or data structures and that can be accessed by general-purpose or special-purpose computer systems. A "network" can be defined as one or more data links that enable electronic data to be transmitted between multiple computer systems and / or multiple modules and / or other electronic devices. When information is transferred or provided to a computer system over a network or another communication connection (hardwired, wireless, or a combination of hardwired and wireless), the computer system may view the connection as a transmission medium. Combinations of the above should also be included within the scope of computer-readable media.

[0087] Furthermore, upon reaching various computer system components, program code in the form of computer-executable instructions or data structures can be automatically transferred from transmission media to computer storage media (or vice versa). For example, computer-executable instructions or data structures received over a network or data link can be buffered in RAM within a network interface module (such as a "NIC") and then ultimately transferred to computer system RAM and / or non-volatile computer storage media at the computer system. Thus, it should be understood that computer storage media can be included in computer system components that also (or even primarily) utilize transmission media.

[0088] Computer-executable instructions may include, for example, a plurality of instructions and data that, when executed by one or more processors, cause a general-purpose computer system, a special-purpose computer system, or a special-purpose processing device to perform a specific function or group of functions. Computer-executable instructions may be, for example, in binary format, an intermediate format such as assembly language, or even source code.

[0089] The disclosure of this application can be practiced in a network computing environment having numerous computer system configurations, including, but not limited to, personal computers, desktop computers, laptop computers, information processors, handheld devices, multiprocessor systems, microprocessor-based or programmable consumer electronics, network personal computers, microcomputers, mainframe computers, mobile phones, personal digital assistants, tablet computers, pagers, routers, switches, and the like. The disclosure of this application can also be practiced in a distributed system environment, in which both local and remote computer systems connected by a network (by hardwired data links, wireless data links, or a combination of hardwired and wireless data links) perform tasks. Thus, in a distributed system environment, a computer system can include multiple component computer systems. In a distributed system environment, program modules can be located in both local and remote memory storage devices.

[0090] The disclosure of this application can also be practiced in a cloud computing environment. A cloud computing environment may be distributed, although this is not required. When distributed, the cloud computing environment can be internationally distributed within an organization and / or have components owned across multiple organizations. In this specification and the following claims, "cloud computing" is defined as a model that provides on-demand network access to a shared pool of configurable computing resources (such as networks, servers, storage, applications, and services). The definition of "cloud computing" is not limited to any of the many other advantages that can be achieved from such a properly deployed model.

[0091] Cloud computing models can be comprised of various characteristics, such as on-demand provisioning of services, extensive and diverse network access, shared resource pools, rapid redeployment flexibility, and scalable services. Cloud computing models can also take the form of various service models, such as Software as a Service ("SaaS"), Platform as a Service ("PaaS"), and Infrastructure as a Service ("IaaS"). Cloud computing models can also be deployed using various deployment models, such as private cloud, community cloud, public cloud, and hybrid cloud.

[0092] Certain embodiments, such as cloud computing environments, may include a system comprising one or more hosts, each of which is capable of running one or more virtual machines. During operation, the virtual machines emulate an operating computing system, supporting an operating system and perhaps one or more other applications. In certain embodiments, each host includes a hypervisor that emulates virtual resources for the virtual machines using physical resources abstracted from the virtual machine's perspective. The hypervisor also provides appropriate isolation between virtual machines. Thus, from the perspective of any given virtual machine, the hypervisor provides the illusion that the virtual machine is interacting with physical resources, even though the virtual machine is only interacting with the appearance of a physical resource (such as a virtual resource). Examples of physical resources include processing capacity, memory, disk space, network bandwidth, media drives, and the like.

[0093] Throughout the specification and claims, certain specific terms are used to refer to particular methods, features, or components. As will be appreciated by those skilled in the art, different persons may refer to the same method, feature, or component by different names. This disclosure is not intended to distinguish between methods, features, or components that differ in name but not function. The figures are not necessarily drawn to scale. Certain specific features and components may be shown exaggerated or in somewhat schematic form, and certain details of conventional components may not be shown or described for the sake of clarity and conciseness.

[0094] Although various exemplary embodiments have been described in detail herein, those skilled in the art will readily appreciate, in light of the present disclosure, that numerous modifications are possible in the exemplary embodiments without materially departing from the concepts disclosed herein. Therefore, any such modifications are intended to be included within the scope of the present disclosure. Similarly, although the disclosure herein contains numerous features, these features should not be inferred to limit the scope of the discloser or any of the appended claims, but are merely intended to provide information directly related to one or more specific embodiments that may fall within the scope of the discloser and the appended claims. Any of the features described from the various disclosed embodiments may be used in combination. In addition, other embodiments of the present disclosure may be devised that fall within the scope of the discloser and the appended claims. Every addition, deletion, and modification made to the embodiments that falls within the meaning and scope of the claims is covered by the claims.

[0095] A set of numerical upper limits and a set of numerical lower limits have been used to describe certain specific embodiments and features. It should be appreciated that, unless otherwise indicated, ranges encompassing any combination of two values ​​are contemplated, such as any combination of a lower limit with any upper limit, any combination of any two lower limits, and / or any combination of any two upper limits. Certain specific lower limits, upper limits, and ranges may be disclosed in one or more claims below. Any numerical value is "almost" or "approximately" an indicated value, and takes into account experimental error and variations that would be expected by one skilled in the art.

[0096] Note that in the above embodiments, the set of components fluidly connected to the shared compressed air distribution network is a limited set of components.

[0097] The present disclosure provides various examples, embodiments, and features, which should be understood to be combinable with other examples, embodiments, or features described herein unless expressly stated otherwise or unless they are mutually exclusive.

[0098] In addition to the above, further embodiments and examples include the following:

[0099] 1. A compressor system comprising: a set of components fluidly connected to a shared compressed air distribution network; and a controller configured to: predict a future demand for the compressor system; determine an initial switching sequence for operation of at least one component of the set of components that meets the predicted future demand; determine a set of switching times for the initial switching sequence; optimize the initial switching sequence based on the set of switching times to form an optimized switching sequence; iteratively determine a set of switching times for the optimized switching sequence and optimize the optimized switching sequence based on the set of switching times until a final switching sequence and a final set of switching times are obtained; and control the operation of the set of components based on the final switching sequence and the final set of switching times.

[0100] 2. The compressor system according to any one of items 1 and 3 to 8 or a combination thereof, wherein the at least one component of the set of components comprises more than one compressor.

[0101] 3. The compressor system according to any one of items 1 to 2 and items 4 to 8 or a combination thereof, wherein the initial switching sequence represents a single sequence of operation of the at least one component.

[0102] 4. The compressor system according to any one of items 1 to 3 and items 5 to 8 or a combination thereof, wherein the at least one component of the set of components comprises one or more dryers.

[0103] 5. The compressor system according to any one of items 1 to 4 and items 6 to 8 or a combination thereof, wherein the at least one component of the set of components comprises one or more valves in the compressed air distribution network.

[0104] 6. The compressor system of any one of items 1 to 5 and items 7 to 8 or a combination thereof, further comprising a plurality of sensors positioned to monitor a plurality of operating parameters of the at least one component.

[0105] 7. The compressor system according to any one of items 1 to 6 and item 8 or a combination thereof, wherein the set of switching times for the initial switching sequence is determined by switching time optimization.

[0106] 8. The compressor system according to any one of items 1 to 7 or a combination thereof, wherein said optimizing the initial switching sequence and said optimizing the optimized switching sequence each comprise removing any portion of the respective sequence that is assigned zero time.

[0107] 9. A controller configured to operate a compressor system having a set of components, the set of components being fluidly connected to a shared compressed air distribution network, the controller comprising: a computer-readable storage medium; and a processor configured to: predict a future demand for the compressor system; determine an initial switching sequence for operation of at least one component of the set of components that meets the predicted future demand; determine a set of switching times for the initial switching sequence; optimize the initial switching sequence based on the set of switching times to form an optimized switching sequence; iteratively determine a set of switching times for the optimized switching sequence, and optimize the optimized switching sequence based on the set of switching times until a final switching sequence and a final set of switching times are obtained; and control the operation of the set of components based on the final switching sequence and the final set of switching times.

[0108] 10. The controller according to any one of items 9 and 10 to 13 or any combination thereof, wherein the initial switching sequence represents a single sequence of operations of the at least one component.

[0109] 11. The controller according to any one of items 9 to 10 and items 12 to 13 or a combination thereof, wherein the set of switching times for the initial switching sequence is determined by switching time optimization.

[0110] 12. The controller according to any one of items 9 to 11 and item 13, or a combination thereof, wherein said optimizing the initial switching sequence and said optimizing the optimized switching sequence each comprise removing any portion of the respective sequence that is assigned zero time.

[0111] 13. The controller according to any one of clauses 9 to 12 or any combination thereof, wherein the switching time optimization comprises calculating a set of switching times according to:

[0112]

[0113] Subj.

[0114]

[0115] p low ≤p(t)≤p high

[0116] Q low ≤Q k (t)≤Q high

[0117] t low ≤t(S on，k , S lo，k )≤t high

[0118] S on，k , S lo，k ∈{0,1}

[0119] 14. A computer-implemented method for controlling a compressor system, the compressor system comprising a group of components fluidly connected to a shared compressed air distribution network, the method comprising: predicting a future demand for the compressor system; determining an initial switching sequence for the operation of at least one component of the group of components that meets the predicted future demand; determining a set of switching times for the initial switching sequence; optimizing the initial switching sequence based on the set of switching times to form an optimized switching sequence; iteratively determining a set of switching times for the optimized switching sequence, and optimizing the optimized switching sequence based on the set of switching times until a final switching sequence and a final group switching time are obtained; and controlling the operation of the group of components based on the final switching sequence and the final group switching time.

[0120] 15. The method according to any one of items 14 and 16 to 20 or any combination thereof, wherein the initial switching sequence represents a single sequence of operations of the at least one component.

[0121] 16. The method according to any one of items 14 to 15 and items 17 to 20 or any combination thereof, wherein the set of switching times for the initial switching sequence is determined by switching time optimization.

[0122] 17. The method according to any one of items 14 to 16 and items 18 to 20 or a combination thereof, wherein said optimizing the initial switching sequence and said optimizing the optimized switching sequence each comprise removing any portion of the respective sequence that is assigned zero time.

[0123] 18. The method according to any one of items 14 to 17 and items 19 to 20 or any combination thereof, wherein the switching time optimization comprises calculating a set of switching times according to:

[0124]

[0125] Subj.

[0126]

[0127] p low ≤p(t)≤p high

[0128] Q low ≤Q k (t)≤Q high

[0129] t low ≤t(S on，k , S lo，k )≤t high

[0130] S on，k , S lo，k ∈{0,1}

[0131] 19. A method according to any one of items 14 to 18 and item 20, or a combination thereof, wherein the initial switching sequence is determined by the predicted future demand by one or more of dynamic programming, analytical dynamic programming (ADP), artificial intelligence (AI), heuristics, branch and bound, and linear programming single solvers.

[0132] 20. The method according to any one of items 14 to 19 or any combination thereof, wherein the predicted future demand represents a predicted future pressure and / or airflow demand for the compressor system.

Claims

1. A compressor system comprising: a set of components fluidly connected to a shared compressed air distribution network; as well as Controller, configured as: Forecast future demand for the compressor system; determining an initial switching sequence for operation of at least one component of the set of components that complies with the predicted future demand; determining a set of switching times for the initial switching sequence; optimizing the initial switching sequence based on the set of switching times to form an optimized switching sequence; Iteratively determining a set of switching times for the optimized switching sequence, and optimizing the optimized switching sequence based on the set of switching times until a final switching sequence and a final set of switching times are obtained; as well as Based on the final switching sequence and the final set of switching times, operations of the set of components are controlled.

2. The compressor system of claim 1, wherein the at least one component of the set of components comprises more than one compressor.

3. The compressor system of claim 1, wherein the initial switching sequence represents a single sequence of operation of the at least one component.

4. The compressor system of claim 1, wherein the at least one component of the set of components comprises one or more dryers.

5. The compressor system of claim 1, wherein the at least one component of the set of components comprises one or more valves in the compressed air distribution network.

6. The compressor system of claim 1, further comprising a plurality of sensors positioned to monitor operating parameters of the at least one component.

7. The compressor system of claim 1, wherein the set of switching times for the initial switching sequence is determined by switching time optimization.

8. The compressor system of claim 1, wherein said optimizing said initial switching sequence and said optimizing said optimized switching sequence each comprise removing any portion of the respective sequence that is assigned zero time.

9. A controller configured to operate a compressor system having a set of components fluidly connected to a shared compressed air distribution network, the controller comprising: computer-readable storage medium; as well as Processor, configured as: Forecast future demand for the compressor system; determining an initial switching sequence for operation of at least one component of the set of components that complies with the predicted future demand; determining a set of switching times for the initial switching sequence; optimizing the initial switching sequence based on the set of switching times to form an optimized switching sequence; Iteratively determining a set of switching times for the optimized switching sequence, and optimizing the optimized switching sequence based on the set of switching times until a final switching sequence and a final set of switching times are obtained; as well as Based on the final switching sequence and the final set of switching times, operations of the set of components are controlled.

10. The controller of claim 9, wherein the initial switching sequence represents a single sequence of operation of the at least one component.

11. The controller of claim 9, wherein the set of switching times for the initial switching sequence is determined by switching time optimization. 12 . The controller of claim 9 , wherein said optimizing said initial switching sequence and said optimizing said optimized switching sequence each comprise removing any portion of the respective sequence that is assigned zero time.

13. The controller of claim 11 , wherein the switching time optimization comprises calculating the set of switching times based on:

14. A computer-implemented method for controlling a compressor system comprising a set of components fluidly connected to a shared compressed air distribution network, the method comprising: Forecast future demand for the compressor system; determining an initial switching sequence for operation of at least one component of the set of components that complies with the predicted future demand; determining a set of switching times for the initial switching sequence; optimizing the initial switching sequence based on the set of switching times to form an optimized switching sequence; Iteratively determining a set of switching times for the optimized switching sequence, and optimizing the optimized switching sequence based on the set of switching times until a final switching sequence and a final set of switching times are obtained; as well as Based on the final switching sequence and the final set of switching times, operations of the set of components are controlled. The method of claim 14 , wherein the initial switching sequence represents a single sequence of operations of the at least one component. The method of claim 14 , wherein the set of switching times for the initial switching sequence is determined by switching time optimization. 17 . The method of claim 14 , wherein optimizing the initial switching sequence and optimizing the optimized switching sequence each comprise removing any portion of the respective sequence that is assigned zero time.

18. The method of claim 16, wherein the switching time optimization comprises calculating the set switching time based on:

19. The method of claim 14, wherein the initial switching sequence is determined by the predicted future demand using one or more of dynamic programming, analytical dynamic programming (ADP), artificial intelligence (AI), heuristics, branch and bound, and linear programming single solvers.

20. The method of claim 14, wherein the predicted future demand represents a predicted future pressure and / or airflow demand for the compressor system.

21. The controller of claim 11 , wherein the switching time optimization comprises calculating the set of switching times based on:

22. The method of claim 16, wherein the switching time optimization comprises calculating the set switching time based on:

Images