Disclosed are methods for cleaning of devices, such as heat exchangers, in particular to methods wherein machine learning systems, such as trained neural networks are used for indicating the fouling status the during the cleaning processes.

An estimation device includes a controller that: acquires a first pipe surface temperature and a second pipe surface temperature, wherein the first pipe surface temperature is an outer surface temperature of a pipe at a precipitate generation position at which a precipitate is adhered to an inner surface of the pipe through which a fluid flows, and the second pipe surface temperature is the outer surface temperature of the pipe at a reference position different from the precipitate generation position; calculates an in-pipe fluid temperature that is a temperature of the fluid at the precipitate generation position based on the second pipe surface temperature; and estimates a thickness of the precipitate based on the in-pipe fluid temperature and the first pipe surface temperature.

A transportation container includes a storage volume. A refrigeration system is contained within the storage volume. The refrigeration system includes a fan configured to ingest exterior air, a condenser coil immediately downstream of the fan, and a vent configured exhaust spent cooling air from the storage volume. A controller is configured to control the refrigeration system. the controller includes instructions configured to cause the refrigeration system to detect an indication of operational deterioration of a transportation container refrigeration system, reverse a rotational direction of an airflow fan such that air is drawn over a refrigeration coil and expelled from the transportation container using the airflow fan, and revert the rotational direction of the airflow fan at a conclusion of a cleaning operation.

The invention provides a method of cooling a flowing fluid in a hydrocarbon flow system using a heat exchange cooler apparatus having at least one cooler conduit; and a sensor system. The method comprises flowing the fluid through the cooler conduit from a cooler inlet to a cooler outlet to cool the fluid and operating a cleaning system to deliver energy (e.g. heat) to the cooler conduit to cause solid materials deposited on the interior of the cooler to be released into the flowing fluid. A cooler sensor data set obtained from the sensor system is compared with a reference data set to verify the performance of the cleaning system.

A system and method for cleaning of heat exchanger tubes including an assembly, an indexer and a communication device provided with specialized software and programming. The indexer includes orthogonally arranged first and second arms. A trolley and sensors are provided on the indexer arms. One or more lances are provided on the trolley to deliver water jets into the openings. Sensors measure displacement as the trolley is moved relative to the heat exchanger's face plate. An operator controls the system from a distance away using the communication device. During setup, the pattern of the face plate is learned and mapped utilizing information from the sensors as one of the inputs. This information is utilized to help navigate the face plate during a subsequent cleaning operation. A kit for retrofitting existing X-Y indexers is also disclosed.

A system and method for cleaning of heat exchanger tubes including an assembly, an indexer and a communication device provided with specialized software and programming. The indexer includes orthogonally arranged first and second arms. A trolley and sensors are provided on the indexer arms. One or more lances are provided on the trolley to deliver water jets into the openings. Sensors measure displacement as the trolley is moved relative to the heat exchanger's face plate. An operator controls the system from a distance away using the communication device. During setup, the pattern of the face plate is learned and mapped utilizing information from the sensors as one of the inputs. This information is utilized to help navigate the face plate during a subsequent cleaning operation. A kit for retrofitting existing X-Y indexers is also disclosed.

A method is provided for evaluating simultaneously the effects of multiple, interdependent heat-exchanger degradation modes for a heat exchanger of a power plant in the context of a series of alternative heat-exchanger remediation strategies. The method includes calculating time-varying predicted future progressions of heat exchanger performance metrics for a plurality of alternative heat-exchanger remediation strategies, and calculating time-varying predicted future progressions of financial metrics describing the accumulated financial benefit of each of the strategies. The calculations may be provided in probabilistic terms. A strategy may then be chosen based, at least in part, on the calculated results.

The present invention relates to an apparatus for circulating balls. An embodiment of the present invention includes: a circulation path for circulating balls and fluid, the circulation path comprising a cooling water tube of a heat exchange unit; a pump arranged in the circulation path for pumping the balls and fluid; and a counter for counting the time and the number of the balls in relation to the circulation of the balls.

A cleaning system according to an exemplary aspect of the present disclosure includes, among other things, a spray cleaning head configured to expel a flow of fluid. The spray cleaning head is moveable along a cleaning path. The system further includes a control unit configured to control movement of the spray cleaning head along the cleaning path automatically. A method is also disclosed.

The following provides a system and method to predict an indicator of tube fouling in a fired apparatus such as a boiler. Historical data can be collected when the tubing is still considered to be clean, and used to build a first model between an indicator of fouling, such as tube skin temperature, and boiler load. The actual measurement of that indicator of fouling can then be compared against the model output, such that the error between the model and measurement is considered an indication of the tube fouling. Moreover, the rate of change of the model error can be used to measure the fouling rate. Next, historical data on the fluid feed quality can be collected and together with the historical error rate change data can be combined to develop a second model. This second model reflects how fluid feed quality variables may affect the fouling rate over time.