Automated pump system installation and commissioning

Abstract

The present disclosure relates to a pump system. This disclosure proposes a method using a machine learning algorithm (MLA) for automated pump installation and / or commissioning is provided. The method comprises: retrieving initial data related to a centrifugal pump, wherein the initial data comprises environmental data and / or a pump profile. Further, automatically determining one or more actions required for installing and / or commissioning the centrifugal pump, based on the initial data using the MLA, and determining whether an autonomous execution of the one or more actions by the centrifugal pump is possible. If the autonomous execution is possible, initiating the autonomous execution of the one or more actions. Additionally or alternatively if the autonomous execution is not possible, indicating that the one or more actions need to be manually executed.

Description

[0001] AUTOMATED PUMP SYSTEM INSTALLATION AND COMMISSIONING

[0002] TECHNICAL FIELD

[0003] This disclosure relates generally to the field of pump systems and more specifically to methods and systems for installing and commissioning centrifugal pumps used in such systems. This encompasses the set-up, calibration, and verification processes involved in ensuring that newly installed pumps or pump systems operate correctly and efficiently according to their intended applications. More particularly, the disclosure pertains to smart pump systems equipped with advanced diagnostics and monitoring capabilities for improved reliability and maintenance.

[0004] BACKGROUND

[0005] Industrial pumps are crucial for the efficient movement of fluids in numerous processes and applications. However, the installation and commissioning of these pumps typically require significant technical expertise, traditionally provided by skilled craftsmen.

[0006] The commissioning of pumps is a critical phase in the installation of pump systems. It involves several steps to ensure that the pump or system not only operates according to the specifications for which it was designed but also integrates seamlessly into its intended operational environment. Traditionally, this process has relied heavily on manual inspections and the use of paper-based guides or manuals, which guide the installer through the setup and calibration phases.

[0007] Despite advancements in technology and the development of digital installation guides and applications that aim to simplify the commissioning process, significant challenges remain. Installers often face the issue of not being able to confirm the perfection of the installation until the system is fully operational. This can lead to scenarios where a pump, initially thought to be installed correctly, exhibits operational inefficiencies or failures once it is put into use. Such inefficiencies often manifest in the form of increased energy consumption and operational costs, stemming from the need to address issues that were not apparent at the time of installation.

[0008] Current digital solutions, while useful, do not adequately address the need for realtime, dynamic feedback during the installation process that can predict and rectify potential operational issues before the system is fully operational. There remains a substantial need for an improved method and system that can guide the installer through a foolproof commissioning process, ensuring optimal operation of the pump from the outset, thus saving time, reducing energy consumption, and minimizing costs associated with post-installation adjustments and repairs.

[0009] SUMMARY

[0010] In view of the above, this disclosure aims to introduce a smart pump system with functionalities capable of helping the installer, obtaining information from the installer, and self-diagnosis during installation to ensure proper installation, as well as notifying the installer in case of detected issues or improper installation. One objective is to improve the convenience of pump installation and commissioning. Another objective is to ensure optimal pump operating conditions from the start. Yet another objective is to decrease the need for specialized installation technicians.

[0011] These and other objectives are achieved by the solution of this disclosure as described in the independent claims. Advantageous implementations are further defined in the dependent claims.

[0012] According to a first aspect of the disclosure, a method using a machine learning algorithm (MLA) for automated pump installation and / or commissioning is provided. The method comprises: retrieving initial data related to a centrifugal pump, wherein the initial data comprises environmental data and / or a pump profile; automatically determining one or more actions required for installing and / or commissioning the centrifugal pump, based on the initial data using the MLA; determining whether an autonomous execution of the one or more actions by the centrifugal pump is possible; if the autonomous execution is possible, initiating the autonomous execution of the one or more actions; and / or if the autonomous execution is not possible, indicating that the one or more actions need to be manually executed.

[0013] This disclosure provides a method using the MLA in the centrifugal pump for automated pump installation and / or commissioning, thereby improving the convenience of pump installation, and decreasing the need for specialized installation technicians. The MLA allows the centrifugal pump to be properly installed, even in a situation where the centrifugal pump is not yet in communication with a command center. The MLA ensures optimal pump operating conditions from the start, thus potentially extending the life of the equipment.

[0014] Optionally, the environmental data may comprise one or more of the following:

[0015] - an ambient temperature;

[0016] - an ambient pressure;

[0017] - an ambient humidity;

[0018] - a physical setup of the centrifugal pump; and one or more environmental regulations or restrictions.

[0019] Notably, the environmental data may refer to any geographical, climatic, and infrastructural characteristics relevant to the installation site that may influence the setup and operational efficiency of the equipment. Possibly, the environmental regulations may be local regulations.

[0020] Optionally, the pump profile may comprise one or more of the following:

[0021] - a type of the centrifugal pump;

[0022] - one or more technical specifications of the centrifugal pump; one or more requirements for the centrifugal pump;

[0023] - one or more operating conditions of the centrifugal pump; and one or more performance curves of the centrifugal pump.

[0024] The pump profile may encompass all key characteristics of the centrifugal pump necessary for its installation and optimal functioning. This may include physical attributes such as size, weight, and configuration; technical specifications like power requirements, connectivity options, and performance capacities; as well as functional capabilities that specify the pump’s intended operations and integration requirements with existing systems.

[0025] Optionally, the method may further comprise: receiving a user feedback indicating a result of the executed one or more actions; automatically determining one or more further actions required for installing and / or commissioning the centrifugal pump, based on the user feedback using the MLA; determining whether an autonomous execution of the one or more further actions by the centrifugal pump is possible; if the autonomous execution is possible, automatically executing the one or more further actions by the centrifugal pump; and / or if the autonomous execution is not possible, indicating that the one or more further actions need to be manually executed.

[0026] Notably, for any actions requiring human intervention, an action prompt may be provided to a user, e.g., the installer. After the prompted actions have been taken, the installer can provide the user feedback. For example, the MLA might ask the installer to take a picture of the centrifugal pump. With a feedback picture, the MLA may be configured to analyze proper alignment based on an analysis of the images. Further actions can be determined based on thereof.

[0027] Optionally, the MLA maybe trained with one or more of the following data:

[0028] - one or more specifications of the centrifugal pump;

[0029] - a user guide of the centrifugal pump;

[0030] - an installation manual of the centrifugal pump;

[0031] - real-case data associated with at least one known failure condition and a respective remedial action;

[0032] - one or more local regulations related to the centrifugal pump;

[0033] - environmental data related to the centrifugal pump; and

[0034] - safety and emergency protocols.

[0035] It may be understood that the MLA has undergone extensive training with a rich dataset encompassing a broad spectrum of operational and regulatory information. As a result of this comprehensive training regimen, the proposed MLA exhibits enhanced responsiveness to both standard operational demands and unforeseen circumstances. It seamlessly integrates real-time data with its extensive training inputs to make informed, reliable decisions that improve operational efficiency, ensure regulatory compliance, and uphold the highest safety standards. Training the MLA may refer to training a model within the MLA, for instance, a neural network. The MLA may encompass both the definition of the model and the process of training the model. The model maybe trained based on the dataset.

[0036] Optionally, the one or more actions and / or further actions may comprise one or more of the following:

[0037] - outputting an instruction to a user;

[0038] - performing an installation check on the centrifugal pump;

[0039] - performing a calibration check on the centrifugal pump;

[0040] - performing a dry run testing on the centrifugal pump;

[0041] - performing a wet run testing on the centrifugal pump;

[0042] - performing a safety and emergency protocol check on the centrifugal pump;

[0043] - starting up the centrifugal pump with a selected algorithm;

[0044] - initiating a self-adaptation of the centrifugal pump;

[0045] - determining a root cause if a failure condition is detected;

[0046] - notifying about a detected failure condition and / or a determined root cause; and

[0047] - generating a document reporting the installation and / or commissioning of the centrifugal pump.

[0048] It maybe understood that in general, the list of actions may comprise at least one of (i) physical installation instructions, (ii) sensor calibration instructions, and (iii) obtaining additional information.

[0049] Optionally, the instruction may comprise one or more of the following: an instruction to visually verify, scan, or take pictures to verify one or more physical parts of the centrifugal pump;

[0050] - an instruction to assemble or disassemble the centrifugal pump; an instruction to mount or dismount one or more physical parts to or from the centrifugal pump; - an instruction to connect or disconnect one or more physical parts to or from the centrifugal pump;

[0051] - an instruction to determine an optimal placement for the centrifugal pump;

[0052] - an instruction to connect a power for the centrifugal pump; and

[0053] - an instruction to turn on / off the centrifugal pump;

[0054] - an instruction to open a water supply or an air vent of the centrifugal pump; and

[0055] - an instruction to verify whether an action has been executed.

[0056] Possibly, the outcomes of unknown issues are fed back to update a model of the MLA to improve the accuracy and effectiveness of the MLA.

[0057] Optionally, the one or more actions and / or further actions may comprise performing an installation check on the centrifugal pump, wherein the method may further comprise: analyzing data from an external source to check whether the centrifugal pump is mechanically correctly installed; and / or analyzing electrical parameters of the centrifugal pump to check whether the centrifugal pump is electrically correctly installed.

[0058] Optionally, the one or more actions and / or further actions may comprise performing a calibration check on the centrifugal pump, the method may further comprise adjusting one or more operational settings of the centrifugal pump to match an intended specification of the centrifugal pump.

[0059] Optionally, the method maybe performed by a processor of the centrifugal pump.

[0060] According to a second aspect of the disclosure, a centrifugal pump is provided. The centrifugal pump comprises a processor configured to perform the method according to the first aspect or any optional implementation form of the first aspect.

[0061] This disclosure further proposes a pump, particularly a smart pump with functionalities capable of helping the user, particularly an installer, obtaining information from the installer, and self-diagnosis during installation to ensure proper installation, as well as notifying the installer in case of detected issues or improper installation.

[0062] Optionally, the processor may be connected to a memory preloaded with a machine learning algorithm, MLA, wherein the processor is configured to execute the MLA.

[0063] Optionally, the centrifugal pump may further comprise a user interface configured to provide an instruction to a user, and / or receive a feedback from a user.

[0064] Possibly, the user interface may include a built-in or externally connected microphone to take voice commands from the user, or a keyboard or a virtual keyboard on a touchscreen to provide instructions to the user and take text feedback from the user. Possibly, the user interface may also have a built-in speaker for providing audio instruction to the user.

[0065] Implementation forms of the centrifugal pump of the second aspect may correspond to the implementation forms of the method of the first aspect described above. The centrifugal pump of the second aspect and its implementation forms achieves the same advantages and effects as described above for the method of the first aspect and its implementation forms.

[0066] According to a third aspect of this disclosure, a computer program product, stored on a non-transitory computer-readable medium, is provided. The computer program product comprises instructions that, when executed by a processor, perform the method according to the first aspect or any optional implementation form of the first aspect.

[0067] All steps that are performed by the various entities described in the present application as well as the functionalities described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. Even if, in the following description of specific embodiments, a specific functionality or step to be performed by external entities is not reflected in the description of a specific detailed element of that entity that performs that specific step or functionality, it should be clear for a skilled person that these methods and functionalities can be implemented in respective software or hardware elements or any kind of combination thereof.

[0068] BRIEF DESCRIPTION OF DRAWINGS

[0069] The above-described aspects and implementation forms will be explained in the following description of specific embodiments in relation to the enclosed drawings, in which

[0070] FIG. 1 shows a method according to an embodiment of this disclosure;

[0071] FIG. 2 shows a centrifugal pump according to an embodiment of this disclosure; and

[0072] FIG. 3 shows a pump commissioning flow according to an embodiment of this disclosure.

[0073] DETAILED DESCRIPTION OF EMBODIMENTS

[0074] Illustrative embodiments of a method using an MLA for automated pump installation and / or commissioning, and a centrifugal pump are described with reference to the figures. Although this description provides a detailed example of possible implementations, it should be noted that the details are intended to be exemplary and in no way limit the scope of the application.

[0075] In this disclosure, an embodiment / example may refer to other embodiments / examples. For example, any description including but not limited to terminology, element, process, explanation, and / or technical advantage mentioned in one embodiment / example is applicable to the other embodiments / examples. The same elements are labeled with the same reference signs and may function similarly or likewise.

[0076] FIG. 1 shows a method 100 using an MLA in a centrifugal pump for automated pump installation and / or commissioning, according to an embodiment of this disclosure. The method 100 comprises a step 101 of retrieving initial data related to a pump, wherein the initial data comprises environmental data and / or a pump profile. The method 100 further comprises a step 102 of automatically determining one or more actions required for installing and / or commissioning the centrifugal pump, wherein the one or more actions are determined based on the initial data using the MLA. Further, the method 100 comprises a step 103 of determining whether an autonomous execution of the one or more actions by the centrifugal pump is possible. If the autonomous execution is possible, the method 100 further comprises a step 104 of initiating the autonomous execution of the one or more actions. Alternatively, if the autonomous execution is not possible, the method 100 further comprises a step 105 of indicating that the one or more actions need to be manually executed.

[0077] Typically, the commissioning process for a pump involves multiple steps as follows:

[0078] 1. Physical examination: Examination of the pump to confirm that it is designed as per design specifications. Generally speaking, the physical examination is done at the factory before sending the pump to the installation facility.

[0079] 2. Installation check: Ensure that installation has been done correctly and the mechanical and electrical connections are properly fitted and functional.

[0080] 3. Calibration: Adjusting the pump’s operational settings to match the intended specification.

[0081] 4. Dry run testing: Operating the pump without fluid to ascertain mechanical and electrical integrity.

[0082] 5. Wet run testing: Introducing the intended fluid into the system and operating the pump to monitor for leaks, vibrations, and abnormal noises, and verifying that the pump delivers the intended flow rate and pressure.

[0083] 6. Safety and emergency protocol check: Evaluating the safety features and emergency shutdown procedures of the pump, as well as local regulation compliance check.

[0084] 7. Final documentation: Creating a comprehensive report and handing over the pump to the operating team.

[0085] The present disclosure focuses on leveraging artificial intelligence (Al), i.e., the MLA, during the commissioning process, in order to limit human interaction. For instance, the MLA may leverage embedded sensors of the pump and preloaded instructions, in order to guide the installer into the correct installment before running or testing / calibration. The MLA may further identify any deviations that could indicate errors and root causes in the centrifugal pump. Automatic adjustment of one or more pump parameters, for instance, to resolve errors or remove root causes, may be implemented using the MLA. This reduces manual turning time and ensures more precise configurations for the pump. In addition, with the help of MLA, the pump can be aware of relevant standards and regulations, and may also be capable of verifying that the commissioning process complies with the relevant standards. Moreover, the MLA may be further configured to compile an installation data log to confirm the proper installation of the pump. This data log may provide a record of the correct commissioning process.

[0086] Notably, the pump discussed in the present application is a centrifugal pump, which operates by converting rotational kinetic energy from a motor into hydrodynamic energy, efficiently moving fluids through the system using a rotating impeller and a volute casing. For ease of description, it may only be referred to as “the pump”, but it is understood to be the centrifugal pump.

[0087] The method 100 shown in FIG. i describes the use of MLA for automated pump installation and / or commissioning. Optionally, the method 100 maybe performed by a processor n of the centrifugal pump 10, as shown in FIG. 2. In a particular example, the centrifugal pump 10 is integrated with a dedicated processing module which is configured to store and perform the MLA. The module comprises the processor 11 in this case.

[0088] The processing module may comprise the processor 11 or generally processing circuitry configured to perform, conduct, or initiate the various operations of the processing module described herein. The processing circuitry may comprise hardware and / or the processing circuitry maybe controlled by software. The hardware may comprise analog circuitry digital circuitry, or both analog and digital circuitry. The digital circuitry may comprise components such as application-specific integrated circuits (ASICs), field- programmable arrays (FPGAs), digital signal processors (DSPs), or multi-purpose processors. The processing module may further comprise memory circuitry, which stores one or more instruction(s) that can be executed by the processor or by the processing circuitry, in particular under the control of the software. For instance, the memory circuitry may comprise a non-transitory storage medium storing executable software code which, when executed by the processor or the processing circuitry, causes the various operations of the processing module to be performed. In one example, the processing circuitry comprises one or more processors and a non- transitory memory connected to the one or more processors. The non-transitory memory may carry executable program code which, when executed by the one or more processors, causes the processing module to perform, conduct, or initiate the operations or methods described herein.

[0089] Optionally, as shown in FIG. 2, the processor 11 is connected to a memory 12 preloaded with an MLA, wherein the processor 11 is configured to execute the MLA. Further optionally, as also shown in FIG. 2, the processor 11 is connected to a user interface 13. The user interface 13 is configured to provide an instruction to a user, and / or receive a feedback from a user. The processor 11 may be further configured to control the user interface 13, and or base its functional processing based on the user feedback.

[0090] The above accordingly proposes a pump 10 with functionalities that are capable of helping the user, particularly an installer, obtaining information from the installer, and self-diagnosis during installation to ensure proper installation, as well as notifying the installer in case of detected issues or improper installation.

[0091] The centrifugal pump 10 with the functionalities is loaded with a combination of (i) installation data (i.e. specific instructions on installing the centrifugal pump 10) and (ii) self-commissioning capabilities, which typically includes the detection / correction of issues.

[0092] The MLA may generally be referred to as an algorithm or learning procedure. The MLA may be stored in the memory 12 and may be executed by the processor 11. The MLA may comprise a set of rules and / or may employ statistical techniques to identify patterns and / or correlations in the initial data (here, the initial data comprises environmental data and / or a pump profile) and may use these patterns and / or correlations to determine a set of actions required for installation and commissioning of the centrifugal pump 10. The MLA may analyze the initial data, learn from the patterns and / or correlations, and may make decisions or predictions without being explicitly programmed for the specific task. For example, the MLA may learn by processing the input data and using it to train a model, wherein the model may comprise a neural network. The model may represent the learned patterns and / or correlations, and the model can be used by the MLA to make decisions or predictions autonomously based on new input data (e.g., user feedback). In an example, the MAL is trained to determine the root cause of an issue and execute corrective measures on the processing module if allowed. Training the MLA may refer to training the model within the MLA. The MLA may accordingly encompass both the definition of the model and the process of training the model.

[0093] According to an embodiment of this disclosure, the MLA has undergone extensive training with a rich dataset encompassing a broad spectrum of operational and regulatory information, which may include one or more of the following:

[0094] Technical Specifications and Equipment Manuals: The MLA may be trained using detailed technical specifications and user guides that describe the pump's components, operational functions, and maintenance requirements. This training ensures that the MLA understands the intricate details of the pump it manages, enabling it to optimize the installation and commissioning procedure effectively.

[0095] Regulatory and Compliance Documents: Understanding and integrating local and international regulatory requirements maybe beneficial for MLA operating in complex environments. The MLA may be trained on a comprehensive array of regulatory documents that dictate operational standards and safety measures specific to the pump’s applications. This may include everything from emission standards to operational safety guidelines, ensuring full compliance and legal operation.

[0096] Real-Case Data and Known Failure Conditions: To enhance its diagnostic and predictive capabilities, the MLA may be trained with historical real-case operational data, including documented instances of equipment failures. This dataset not only includes descriptions of failure scenarios, but also details the remedial actions taken. This provides the MLA with the ability to quickly diagnose issues and propose proven solutions, thereby minimizing downtime and enhancing reliability.

[0097] Environmental Data Inputs: Environmental conditions can significantly affect equipment performance. The MLA maybe trained to interpret and react to a variety of environmental inputs such as temperature, humidity, and other atmospheric conditions, which are critical to adapting operations in real time to maintain efficiency and safety.

[0098] Safety and Emergency Protocols: One of the MLA's training components involves safety and emergency response protocols. By integrating detailed emergency procedures into the MLA’s knowledge base, the system is equipped to initiate immediate and appropriate responses to potential safety incidents, ensuring the protection of both the equipment and personnel.

[0099] Local Regulations and Standards: The training also includes detailed information on local regulations that affect equipment operation. This aspect of training ensures that the MLA can adapt its operations to meet local standards, which may vary significantly across different regions.

[0100] As a result of this comprehensive training based on one or more of the above, the MLA may exhibit enhanced responsiveness to both standard operational demands and unforeseen circumstances. It may seamlessly integrate real-time data with its extensive training inputs to make informed, reliable decisions that improve operational efficiency, ensure regulatory compliance, and uphold the highest safety standards. Its ability to learn from real-case scenarios and adapt to a diverse range of operational conditions and regulations makes it an invaluable asset in managing complex equipment systems.

[0101] The initial data provides the foundational input that the MLA uses to make decisions about the installation and commissioning process. This may include determining the optimal configurations and adjustments needed to successfully install and commission the pump. The MLA’s ability to process and analyze these parameters effectively can directly influence the efficiency, accuracy, and success of the installation and commissioning.

[0102] For instance, different types of centrifugal pumps might require different installation parameters such as size, weight, power requirements, and environmental conditions. The initial data allows the MLA to customize its approach based on these specific requirements, ensuring that the installation is performed with precision tailored to each piece of equipment.

[0103] In addition, with access to detailed initial parameters, the MLA can perform predictive analysis to foresee potential issues and proactively suggest adjustments. This can prevent installation errors and reduce downtime, which is crucial in industries where equipment setup times are critical.

[0104] The initial data may comprise the environmental data and / or the pump profile. Notably, the environmental data refers to any geographical, climatic, and infrastructural characteristics relevant to the installation site that may influence the setup and operational efficiency of the equipment. This data includes but is not limited to, temperature, humidity, pressure, altitude, regulatory constraints, and existing infrastructural compatibilities.

[0105] For instance, the environmental data comprises one or more of the following:

[0106] - an ambient temperature;

[0107] - an ambient pressure;

[0108] - an ambient humidity;

[0109] - a physical setup of the centrifugal pump; and one or more local environmental regulations or restrictions.

[0110] For instance, the physical setup of the system may include one or more of the following information:

[0111] - the number of valves;

[0112] - a physical location of the pump (e.g., in the basement or another place);

[0113] - whether the pump is included in a booster system (multiple installations of pumps), if yes the amount of pumps; and the size of the system where the pump is installed, e.g., the size of the area the system spans or covers (such as multiple floors of a building, a large industrial facility, etc.).

[0114] The pump profile may encompass all key characteristics of the pump necessary for its installation and optimal functioning. This includes physical attributes such as size, weight, and configuration; technical specifications like power requirements, connectivity options, and performance capacities; as well as functional capabilities that specify the pump’s intended operations and integration requirements with existing systems.

[0115] For instance, the pump profile comprises one or more of the following:

[0116] - a type of the pump;

[0117] - one or more technical specifications of the pump; one or more requirements for the pump;

[0118] - one or more operating conditions of the pump;

[0119] - one or more performance / operational curves of the pump.

[0120] The pump profile may also encompass performance / operational curves. These could, e.g., include one or more of the following:

[0121] - Q-H curve, which refers to the relationship between head (H) and flow rate (Q);

[0122] - efficiency curve, which shows the pump’s efficiency at various operating points;

[0123] - power / energy curve, which relates the pump’s power consumption (usually in kilowatts) to the flow rate;

[0124] - NPSHR (Net Positive Suction Head Required), which indicates the minimum suction pressure required to prevent cavitation; and

[0125] - pump family curve.

[0126] Notably, the Q-H curve illustrates how head changes with varying flow rates. Engineers may use it to select the right pump for specific applications.

[0127] Efficiency curve helps to determine the most efficient flow rate for a given pump. Ideally, it is desired to operate the pump near its peak efficiency point to minimize energy consumption and maximize performance. Power / energy curve helps assess the energy requirements for different flow rates.

[0128] Operating the pump at the right flow rate ensures optimal energy usage.

[0129] Cavitation occurs when the pressure at the pump inlet drops too low, causing vapor bubbles to form and collapse, damaging the impeller and reducing efficiency. The NPSHR value depends on the pump design and operating conditions.

[0130] Some manufacturers provide a family of curves for a specific pump model. These curves show variations in impeller diameter, speed, or other design parameters. Engineers can choose the right impeller size or speed based on system requirements.

[0131] FIG. 3 illustrates a flow chart of how the MLA is used for automated pump installation and / or commissioning, according to an embodiment of this disclosure. The embodiment of FIG. 3 is based on the embodiments of FIG. 1 and FIG. 2. The illustrated actions performed by the MLA maybe executed by the processor 11 shown in FIG. 2.

[0132] First of all, it may be understood that before installation / commissioning of the pump, the MLA may store information about the site of the installation and if prompted by the installer, provide the following information:

[0133] - location of the installation site;

[0134] - the pump type being installed;

[0135] - contact person (ex. Site manager) at the installation site; and

[0136] - restrictions and / or limitations on the installation of the pump (e.g., require electrician, etc.).

[0137] It may be worth mentioning that how this information is collected is not limited in the present disclosure. Possibly, it may have been provided by the client during a sizing process.

[0138] If applicable, dismounting an existing pump may be required. The MLA may store information about the dismounting of the currently replaced pump, and if prompted by the installer, provide instructions, such as:

[0139] - disconnecting the power, PLC, and other external wiring to the replaced pump; - discharge pressure of the replaced pump;

[0140] - drain water from the replaced pump or system; and

[0141] - instructions on dismounting, and separating the replaced pump from the system.

[0142] In other words, the MLA provides the instructions to be performed by the installer, and once the installer has executed the action, he or she can prompt the MLA to provide the next instruction.

[0143] How this information is first collected is not limited in this disclosure, and may have been provided by the client during the sizing process, or alternatively, by accessing the dismounting instruction associated with the pump.

[0144] Once the existing pump (if any) is dismounted, the installer then proceeds to assemble the pump 10 and install it, which may include the steps:

[0145] - the MLA prompts the installer to visually verify, scan, or take pictures to verify that the correct parts have been delivered and are not damaged;

[0146] - the MLA prompts the installer to visually verify, scan, or take pictures to verify that the pipework is suitable;

[0147] - assembling the pump 10 and mounting the pump 10;

[0148] - the MLA is configured to perform safety & standard checks, such as selfverifying that the gaskets are in place, the coupling between the motor and pump-end is fixed correctly, etc.;

[0149] - the MLA is configured to provide instructions to the installer to mount external sensors and PLC as required and confirm proper installation;

[0150] - the MLA is configured to provide instructions to the installer to connect power and confirm;

[0151] - the MLA is configured to provide instructions to the installer to turn on the pump 10 in idle mode;

[0152] - the MLA may be configured to self-assess proper connection to the cloud and the command center;

[0153] - the MLA may be configured to initiate a self-test of the processing module, e.g. its hardware, etc. to confirm proper working, in conjunction with instructions received by the command center; - the MLA may be configured to transfer a confirmation that the right pump is installed in the right site, this is particularly useful for an installation site requiring multiple installations of pumps;

[0154] - the MLA may be configured to retrieve and update the pump software to the latest relevant software; and

[0155] - the MLA may prompt the installer to open the water supply or air-vent the pump, and self-assess that such action has been taken.

[0156] Once the pump 10 has been assembled and installed, the steps before concluding the commissioning include:

[0157] - the MLA is configured to start up the pump 10 slowly by running selected algorithms in over-clock mode to ensure that the pump 10 functions properly;

[0158] - the MLA can also initiate self-test / adaptation, such as:

[0159] • Changing the RPM

[0160] • Confirm 2-way connection with SCADA / BMS system

[0161] • Collect operational data and adjust and / or update software / pump settings to a desired state

[0162] • If any issues are detected, determine the root cause and notify the command center / installer or self-rectify the issue; and

[0163] - the MLA can also initiate sensor calibration checks and provide instructions to the installer for rectifying calibration issues.

[0164] Details of the automated installation and / or commissioning procedure are now explained with reference to FIG. 3. At step 301, the pump 10 is turned on. The pump 10 may comprise a processing module, e.g., including the processor 11 as shown in FIG. 2, with a memory 12, storing an MLA that has been pre-loaded. The MLA is designed to handle various stages of the pump installation process by learning from data to automate decisions and actions that would traditionally be performed manually by technicians.

[0165] At step 302, the MLA determines a set of actions required to achieve a desired state.

[0166] The MLA analyzes the initial data, which might include specifications of the pump 10, environmental conditions at the installation site, and predefined installation parameters. This step ensures that the MLA understands the context in which the pump 10 will operate. Based on the analysis, the MLA makes decisions about the best practices for installation. This could involve choosing the appropriate tools, methods, and sequences for assembly based on the model's training on past installations.

[0167] In some embodiments, the memory 12 as shown in FIG. 2 comprises a list of actions to be taken to complete the installation. In general, the list of actions comprises at least one of (i) physical installation instructions, (ii) sensor calibration instructions, and (hi) obtaining additional information.

[0168] Physical installation instructions include mechanical and electrical installation instructions. Relevant actions may include optimal placement, securement to the base, alignment with piping, and connection to the power source.

[0169] The memory 12 may also store action regarding the installation and calibration of the sensors. Further, some of the actions may prompt the installer to provide information about the installation site, such as the number of valves, the ambient temperature, etc.

[0170] For instance, the list of actions comprises one or more of the following:

[0171] - outputting an instruction to a user;

[0172] - performing an installation check on the pump 10;

[0173] - performing a calibration check on the pump 10;

[0174] - performing a dry run testing on the pump 10;

[0175] - performing a wet run testing on the pump 10;

[0176] - performing a safety and emergency protocol check on the pump 10;

[0177] - starting up the pump with a selected algorithm;

[0178] - initiating a self-adaptation of the pump 10;

[0179] - determining a root cause if a failure condition is detected;

[0180] - notifying about a detected failure condition and / or a determined root cause;

[0181] - generating a document reporting the installation and / or commissioning of the pump 10.

[0182] At step 303, the MLA determines whether an autonomous execution of the one or more actions by the pump 10 is possible. For any actions requiring human intervention, the MLA provides an action prompt to the installer at step 304.

[0183] Optionally, the instruction comprises one or more of the following: an instruction to visually verify, scan, or take pictures to verify one or more physical parts of the pump 10;

[0184] - an instruction to assemble or disassemble the pump 10; an instruction to mount or dismount one or more physical parts to or from the pump 10 ; an instruction to connect or disconnect one or more physical parts to or from the pump 10 ;

[0185] - an instruction to determine an optimal placement for the pump 10;

[0186] - an instruction to connect a power for the pump 10;

[0187] - an instruction to turn on / off the pump 10;

[0188] - an instruction to open a water supply or an air vent of the pump 10; and

[0189] - an instruction to verify whether an action has been executed.

[0190] For example, the MLA might ask the installer to take a picture of the pump 10. Leveraging computer vision technology and image-processing algorithms, the processing module is configured to analyze proper alignment based on an analysis of the images taken by the camera.

[0191] In another example, during the initial setup, the MLA would prompt the installer or technician for the calibration procedures for each sensor type. This process can involve instructing the user to place reference weights, apply known pressures, or simulate standard temperature conditions to check sensor accuracy.

[0192] In another embodiment, the MLA might prompt the installer to take pictures or input information relating to the environment of the pump 10 or the pump system itself (such as the number of valves, etc.). It should be noted that the provided actions are not only for the installation process by itself, but the actions may also be input for the MLA to achieve the desired state configuration (see below in step 310).

[0193] It should be understood that the way in which the installer provides the information to the MLA is not limited. For example, the pump may comprise a human-machine interface (HMI) with a keyboard in which the installer can input the information, or be connected to a handheld device of the installer, or using LLM.

[0194] If the autonomous execution by the pump 10 is possible, the MLA initiates the autonomous execution of the one or more actions. For instance, to perform an installation check on the pump, the MLA may analyze data from an external source to check whether the pump is mechanically correctly installed; and / or analyze the electrical parameters of the pump to check whether the pump is electrically correctly installed.

[0195] To perform a calibration check on the pump, the MLA may adjust one or more operational settings of the pump to match an intended specification of the pump 10.

[0196] At step 305, once the installation of the pump 10 is completed, test runs (dry and wet) are made. These are part of the normal commissioning process and well known in the art, and thus will not be covered in detail herein.

[0197] At steps 306-309, during the test runs, issues or anomalies can be detected by the MLA. In some embodiments, the MLA can also detect the root cause of the anomaly as shown in step 307, and execute remedial action on its own if it can (see step 309). Otherwise, the MLA notifies the command center or the installer of the anomaly at step 308.

[0198] Possibly, the MLA may be configured to automatically diagnose the root cause of a failure condition of the pump 10 based on data received from the pump 10 (e.g., from one or more sensors of the pump 10). The MLA determines whether the pump 10 is able to take corrective measures automatically, and proceeds accordingly based on the outcome of the determination. In particular, if the autonomous correction is possible, the pump 10 automatically executes the one or more remedial actions identified by the MLA. If the autonomous correction is not possible, the MLA is configured to notify the command center that involvement of the installer is required.

[0199] In an optional embodiment, the MLA can detect a deviation from the standard operating parameters. It can thus detect the root cause and if the root cause allows it, it can further take corrective or remedial actions at the pump 10.

[0200] Optionally, the one or more remedial actions may comprise adjusting one or more operational parameters controlling the pump 10. The MLA may be configured to instruct the autonomous execution of the identified one or more remedial actions by the pump 10, and for instance, instruct the automatically adjustment of the one or more operational parameters of the pump 10.

[0201] At step 310, once the test runs are over, the commissioning enters the last stage in final calibration.

[0202] According to another embodiment of this disclosure, a computer program product, stored on a non-transitory computer-readable medium, is provided. The computer program product comprises instructions that, when executed by a processor, perform the method according to the previous embodiments related to the use of the MLA, for instance, the embodiments shown in FIG. 1 and FIG. 3.

[0203] To summarize, by preloading a processor 11, e.g., a processing module, of the pump 10 with digital instructions for the pump installation, this disclosure improves the convenience of pump installation (reduced mistakes and downtime). The proper installation is possible even in a situation where the pump is not yet in communication with the command center, this is particularly useful when the pump is installed in a remote area or a location with no connection readily available. This disclosure further ensures optimal pump operating conditions from the start, extending the life of the equipment. In addition, it decreases the need for specialized installation technicians.

[0204] Notably, the centrifugal pump 10 discussed in this disclosure may be a smart pump, e.g., a pump integrated with processing capabilities, sensors, software, and connectivity features that is able to perform automated monitoring and control, and particularly benefits from the above-described advantages provided by the processor n.

[0205] The present disclosure has been described in conjunction with various embodiments as examples as well as implementations. However, other variations can be understood and effected by those persons skilled in the art and practicing the claimed matter, from the studies of the drawings, this disclosure, and the independent claims. In the claims as well as in the description the word “comprising” does not exclude other elements or steps and the indefinite article “a” or “an” does not exclude a plurality. A single element or other unit may fulfill the functions of several entities or items recited in the claims. The mere fact that certain measures are recited in the mutually different dependent claims does not indicate that a combination of these measures cannot be used in an advantageous implementation.

Claims

Claims1. A method (100) using a machine learning algorithm, MLA, for automated pump installation and / or commissioning, the method (100) comprising: retrieving (101) initial data related to a centrifugal pump, wherein the initial data comprises environmental data and / or a pump profile; automatically (102) determining one or more actions required for installing and / or commissioning the centrifugal pump, based on the initial data using the MLA; determining (103) whether an autonomous execution of the one or more actions by the centrifugal pump is possible; if the autonomous execution is possible, initiating (104) the autonomous execution of the one or more actions; and / or if the autonomous execution is not possible, indicating (105) that the one or more actions need to be manually executed.

2. The method (100) according to claim 1, wherein the environmental data comprises one or more of the following:- an ambient temperature;- an ambient pressure;- an ambient humidity;- a physical setup of the centrifugal pump; and one or more local environmental regulations or restrictions.

3. The method (100) according to claim 1 or 2, wherein the pump profile comprises one or more of the following:- a type of the centrifugal pump;- one or more technical specifications of the centrifugal pump; one or more requirements for the centrifugal pump;- one or more operating conditions of the centrifugal pump; and one or more performance curves of the centrifugal pump.

4. The method (100) according to one of the claims 1 to 3, comprising: receiving a user feedback indicating a result of the executed one or more actions;automatically determining one or more further actions required for installing and / or commissioning the centrifugal pump, based on the user feedback using the MLA; determining whether an autonomous execution of the one or more further actions by the centrifugal pump is possible; if the autonomous execution is possible, automatically executing the one or more further actions by the centrifugal pump; and / or if the autonomous execution is not possible, indicating that the one or more further actions need to be manually executed.

5. The method (100) according to one of the claims 1 to 4, wherein the MLA is trained with one or more of the following data:- one or more specifications of the centrifugal pump;- a user guide of the centrifugal pump;- an installation manual of the centrifugal pump;- real-case data associated with at least one known failure condition and a respective remedial action;- one or more regulations related to the centrifugal pump;- environmental data related to the centrifugal pump; and- safety and emergency protocols.

6. The method (100) according to one of the claims 1 to 5, wherein the one or more actions and / or further actions comprise one or more of the following:- outputting an instruction to a user;- performing an installation check on the centrifugal pump;- performing a calibration check on the centrifugal pump;- performing a dry run testing on the centrifugal pump;- performing a wet run testing on the centrifugal pump;- performing a safety and emergency protocol check on the centrifugal pump;- starting up the pump with a selected algorithm;- initiating a self-adaptation of the centrifugal pump;- determining a root cause if a failure condition is detected;- notifying about a detected failure condition and / or a determined root cause; andgenerating a document reporting the installing and / or commissioning of the centrifugal pump.

7. The method (100) according to claim 6, wherein the instruction comprises one or more of the following: an instruction to visually verify, scan or take pictures to verify one or more physical parts of the centrifugal pump;- an instruction to assemble or disassemble the centrifugal pump; an instruction to mount or dismount one or more physical parts to or from the centrifugal pump;- an instruction to connect or disconnect one or more physical parts to or from the centrifugal pump;- an instruction to determine an optimal placement for the centrifugal pump;- an instruction to connect a power for the centrifugal pump; and- an instruction to turn on / off the centrifugal pump;- an instruction to open a water supply or an air vent of the centrifugal pump; and- an instruction to verify whether an action has been executed.

8. The method (ioo) according to claim 6 or 7, wherein when the one or more actions and / or further actions comprise performing an installation check on the centrifugal pump, wherein the method (100) further comprises: analyzing data from an external source to check whether the pump is mechanically correctly installed; and / or analyzing electrical parameters of the pump to check whether the pump is electrically correctly installed.

9. The method (100) according to one of the claims 6 to 8, wherein when the one or more actions and / or further actions comprise performing a calibration check on the centrifugal pump, the method (100) further comprises: adjusting one or more operational settings of the centrifugal pump to match an intended specification of the centrifugal pump.

10. The method (100) according to one of the claims i to 9, wherein the method (100) is performed by a processor of the centrifugal pump.

11. A centrifugal pump (10) comprising: a processor (11) configured to perform the method (100) according to one of the claims 1 to 10.

12. The centrifugal pump (10) according to claim 11, wherein the processor (11) is connected to a memory (12) preloaded with a machine learning algorithm, MLA, wherein the processor (11) is configured to execute the MLA.

13. The centrifugal pump (10) according to claim 11 or 12, further comprising: a user interface (13) configured to provide an instruction to a user, and / or receive a feedback from a user.

14. A computer program product, stored on a non-transitory computer-readable medium, comprising instructions that, when executed by a processor (11), perform the method (100) according to one of the claims 1 to 10.

Images