METHOD AND ARRANGEMENT FOR SPREADER MAINTENANCE

The present invention relates to a spreader adapted for lifting freight containers. In particular the invention relates to maintenance arrangements and procedures of spreaders.

Background of the Invention

In the field of container transport, major efforts are continuously being made to reduce port stops and the time needed for loading/unloading of ships, trains and road units. A priority in these efforts is reducing the number of unplanned stops due to the malfunction of the equipment used in the loading/ unloading procedures.

Freight containers are conventionally handled by engaging a lifting device, a spreader, to the top of one or more container. The spreader is connected to a crane by a lifting cable, which lifts or lower the spreader. A spreader is basically a metal frame with four or more engaging elements, adapted to interact with corresponding engaging elements on the top of containers. The most common type of engaging elements are so-called twistlocks. The twistlocks are engaged in twistlock apertures on the containers and rotated to give a fast and secure locking, capable of carrying very high loads.

Modern spreaders are to a high degree automated, and comprise systems for automatic guiding of the twistlocks to the twistlocks apertures on the containers, automatic locking, and automatic adjustment of the distances between containers if more than one container is lifted by the same frame. In addition a number of control and safety systems are involved, for example giving an alarm and hindering the lift, if a twistlock does not lock. The parts of the spreader are typically hydraulically- and/or electrically-operated. The control and safety systems comprise a wide range of sensors, including position sensors, mechanical switches, optical sensors etc.

The high degree of automation and the high degree of exposure to a sometimes extremely harsh environment makes regular maintenance of the spreaders necessary.

The spreaders are operated under very varying conditions, for example from arctic climate to extreme heat. In some cases the environment are such that the wear of certain parts can be accelerated. In other cases the environment causes the hydraulic systems to require more frequent maintenance. In addition the use of spreader, the condition of the crane etc. may also be an influence on maintenance schedules. Therefore, the intervals between maintenance and scheduled replacement of parts may differ considerably from spreader to spreader. At the same time downtime on spreaders is extremely costly. This is especially so if the crane is still in operation. Therefore should all types of service and maintenance preferably be done when the spreader can be taken off the crane without interrupting operations.

US 7,031,883 describes a spreader containing a processor and a data storage device for storing data from sensors on the spreader. This arrangement gives easy access to data that can be used for diagnostic operations and/ or preventive maintenance.

Summary of the Invention

The objective problem is to provide a method and an arrangement for the operation a spreader for reducing the downtime of the spreader.

The problem is solved by the method as defined in claim 1 and the system as defined in claim 10.

The present invention provides a method and arrangement of operating a spreader, the method comprises the steps of:

- monitoring signals from a plurality of sensors and/ or input output units on the spreader; - relating the signals to a parameter or set of parameters, wherein the parameters are associated to events in the operation of the spreader and/or the condition of parts and/or the performance of functions;

- generating and storing at least one trend log, each trend log associated to one defined parameter or a defined set of parameters and thereby to an operation and/or part and/or function of the spreader;

- predicting a future behaviour of a part and/ or function by comparing a trend log associated to a part and/or function with associated reference data.

According to one embodiment the prediction relates to a failure of a part or a function, and to providing an indication if the analysis shows that a failure is approaching. The method may in addition estimate an expected time or number of operations before the failure occur, and/ or give an estimate of the how probable it is that a function or part will operate without failure until the time of a scheduled maintenance.

Thanks to the inventive system and method it is possible to replace a worn, damaged or liable to fail part or perform other types preventive maintenance on the spreader before a failure occurs. This will significantly reduce the downtime of the spreader.

One advantage of the present invention is that references values can be made dynamic, and the reference values and/or the choice of what parameters to monitor in order to supervise a part or function can be a result of automatically analysed trend logs. Embodiments of the invention are defined in the dependent claims. Other objects, advantages and novel features of the invention will become apparent from the following detailed description of the invention when considered in conjunction with the accompanying drawings and claims.

Brief Description of the Drawings

The invention will now be described in detail with reference to the drawing figures, in which:

Fig. 1 is a schematic view of a spreader provided with the diagnostic/ prognostic system according to the invention;

Fig. 2 is a flowchart over the method according to the invention;

Fig. 3 is a graph illustrating an extrapolation process;

Fig. 4 is a graph illustrating a prediction process;

Fig. 5 is a schematic illustration of a twistlock; and,

Fig. 6 is a schematic illustration of the logic of the present invention.

Detailed Description of the invention

The principle of the present invention will be described with reference to the schematic illustrations of FIG. la-b. A spreader 100 is connected via cables to a crane (not shown) . Through twistlocks 115 the spreader can releasably engage with a container 102, as illustrated in FIG Ia. The spreader 100 is provided with a diagnostic and prognostic system 105 that operates autonomously from the control software which controls the operation of the spreader. The diagnostic and prognostic system 105 comprises a plurality of sensors or input/output (I/O) units 110 providing information of the status and operation of a plurality of different parts and functions of the spreader 100, for example the operation of a twistlock 115. The information could, for example, include a time stamp, that is a record of the time each movement of the twistlock takes, the pressure or current required for each movement and the signal level and stability of the sensors monitoring the twistlock. A number of sensors 110 can be utilised for monitoring the same function/ part. The diagnostic and prognostic system preferably operates on the same computer as, but independently from, the software operating the machine. It can collect data, analyse it, update databases and produce forecasts regarding possible faults. The forecasts can be updated using data collected by the diagnostic and prognostic system.

Monitoring and time stamping of external events, e.g. wind speed and direction, hoisting speed of the crane vertically and horizontally, etc may also take place and the values may be stored in a trend log or an event log so that it possible to see if abnormal operating parameters are related to specific external events.

The sensors 110 are connectable to a control unit 120, for example a computer dedicated for operating, and mounted on, the spreader 100. An embodiment of a control unit 120 is schematically illustrated in FIG Ib. In the control unit 120 the signals from the sensors 110 are received by a receiving module 125. The receiving module is connectable to a diagnostic module 127 adapted to interpret the signals from the sensors 110. Signals from a sensor may for example be related by the diagnostic module 127 to a specific position, pressure etc. A sequence of signals from one or a plurality of sensors is interpreted by the receiving module as a

procedure for which, for example, the time from start to completion, is monitored. Thus a number of predefined parameters which are monitored by the control unit 120 are defined in the receiving module 125, each parameter corresponding to a receiving sub-module.

The diagnostic module 127 is via a trend selection module 130 connectable with a trend database 135, adapted to store values of parameters, or representations of them, and a time stamp and/or operation number associated to each trend database entry of the value of a parameter. The trend database 135 may during operation store a plurality of trend logs (where a trend log comprises a series of entries of the recorded value of a parameter and the time and/ or the operation number of the operation in progress when each entry was measured) , based on the selected parameters provided by the trend selection module 130. The control unit 120 further comprises a prognostic module 140, connectable to the trend database 135, the trend selection module 130 and a reference database 145. The prognostic module 140 is adapted to instruct the selection module 130 which parameters to select and create trend logs from. The reference database 145 comprises information relating to the parameters, for example information on acceptable ranges for the different parameters (e.g. values of acceptable operating pressures) and information on expected behaviour of certain parameters over time (e.g. information that twistlock operating pressures may rise over time due to increased friction as lubricants are worn away) . Information may be pre-stored in the database and/ or loaded from an external source. This information may include generic information about the spreader (type, model etc) which typically may be provided by the manufacturer of the spreader. Part of information stored in the reference database 145 may be data generated from the values for parameters produced by the diagnostic and prognostic system and based on the actual spreader.

The prognostic module 140 is arranged to compare trend logs for a parameter, or a set of parameters, retrieved from the trend database 135, with information retrieved from the references database 145. By comparing the trend log of a parameter or a set of parameters with the information provided by the reference database 145, the prognostic module 140 can make a prediction of, for example, the remaining time or remaining number of operations before replacement of a part or maintenance is

due. Methods for performing predictions will be further discussed below. The prognostic module 140 is connectable to an output module 150, which preferably is adapted to present the prediction to, for example, an operator or maintenance organisation responsible for maintenance. The prediction may for example be presented on a GUI or in form of a message transmitted via air interface to a central maintenance system or to a personal device as a mobile phone /communicator.

The diagnostic and prognostic system 105 may also comprise, in the control unit 120, a number of other modules providing the operator and/ or the organisation maintenance with information on the status of the spreader. Such modules may include, but are not limited to, a status module 155 and an alarm module 160, both connectable with the receiving module 125 and the reference database 145. The status module 155 is adapted to provide information on the current status of all parameters, or a selection of parameters. The alarm module 160 is adapted to issue immediate alarms and/ or arrest the operation if for example the value of a certain parameter is outside an allowed range, the extent of the allowed range being provided from the reference module 145.

In the prognostic method according to the present invention, described with reference to the flowchart of FIG. 2, references data, locally stored on the spreader, is used in combination with trend logs based on measured parameters, to produce predictions regarding replacement of parts, repairs, adjustments and other maintenance issues. The method comprises the steps of:

205: "Monitoring signals from sensors" -Monitoring signals from a plurality of sensors and/or input output units on the spreader.

210: "relating signals to parameters" - Relating the signals to a set of parameters. The parameters correspond to events in the operation of the spreader and/or conditions of parts or the performance of functions.

220: "Generating/ storing trend logs" - Generating and storing trend logs.

Each trend log is associated to one defined parameter or a defined set of parameters. The trend log also comprises a time stamp and/or a sequence number (for example an operation number which is

incremented by one each time the operation is performed), enabling a monitoring of the development of parameters over time and/ or over a number of consecutive operations. The trend logs are stored in a local database.

225: "Retrieve trend log/ref data" - A trend log is retrieved from the trend log database and corresponding reference data is retrieved from a reference database.

230: "Compare trend log and ref data to determine prediction" - The trend log is compared with the reference data to determine a prediction associated with the parameter or set of parameters. The prediction is typically an indication of an approaching failure of a part or function. Additionally the prediction may include an expected time or expected number of operations before failure.

235: "Prediction cause for action?" -Based on the prediction it is determined if an action should be recommended/ ordered and reported to an operator or maintenance organisation. If no action needs to be recommended or ordered then the monitoring of steps 205-235 continues.

240: "Issue action order" - An action order is issued; resulting in an output which can be directed to an operator, maintenance organisation, or the like in the form of, for example, a warning, or a rescheduling of maintenance.

245: "Restore trend log to initial value" - After an action is performed the trend log of the associated parameter/ parameters is typically set to an initial value.

During operation of the spreader steps 205-235 are repeated to give a continuous supervision and evaluating of the status of the spreaders.

According to one embodiment of the invention the comparing of step 230 is based on current signal statuses in combination with reference information and at least one stored trend log. By looking at values over time or operations or external events

(as recorded in the trend logs) and comparing with statistical values or reference values, the rules determine if the behaviour of the machine will lead to a situation requiring service and this can be reported prior to the operator being seriously affected by the situation, thus allowing service to be planned in order to minimise downtime. The reference information is in this case one or more references values characteristic for the monitored parameter, the values typically indicating the need for replacement or maintenance. By using the values of the trend log and extrapolating in time, as schematically illustrated in FIG. 3, an estimate is given of the time and/ or number of operations remaining until a probable malfunction. In the figure the solid line represents the reference value, the Xs the trended values (that is historical values recorded in the actual devices trend log) and the dashed line the extrapolation. For simplicity the extrapolation is illustrated as linear, but extrapolations using curves, for example exponential expressions, are equally plausible. Different extrapolation methods are well known in the art. Indicated in the figure is an example of an extrapolated measure, a time before the reference value is reached, for example indicating an estimated time before failure, tbf.

In another embodiment of the invention the fact that the spreader may be operated in very varying conditions, giving essentially different exposure to wear, for example, is taken into account. The reference information associated to one parameter includes a plurality of different scenarios. This information can for example be stored as a plurality of curves with time or number of operations as one axis, and the parameter as the other, each curve representing one scenario. By comparing with the trend log, the reference curve that best represent the current conditions is selected, and used to give a prediction. One example is given in FIG. 5, wherein the curves A, B and C represent the wear of a part in harsh, medium and light environment, respectively. By comparison with the trended values (X) it is determined that curve A is most representative and this is subsequently used to determine the expected lifetime of the part.

Based on the prediction produced in step 240, actions could be automatically recommended/ ordered and presented to an operator of the spreader and or maintenance organisation (step 245). According to one embodiment the reference database also stores information regarding scheduled maintenance. By comparing the predictions for one or more parameters with the time remaining to the next

scheduled maintenance it can be determined if the maintenance should be rescheduled. Alternatively on approaching the time for one scheduled maintenance a list can be produced of events that are predicted to occur in the time interval to the next scheduled maintenance and preventive actions can be taken.

The analysis performed in steps 240-245 may also comprise more elaborate predictions and recommendations. For example the predictions may be of the form "15% chance of failure of component A in 10 days", "20% chance of reduced functionality in process Z in 1000 operations". Through an expert system it can be automatically determined for example if the cost of an extra maintenance should be taken or if the risk involved in waiting for a scheduled maintenance is acceptable.

A basic information content of the reference database is typically provided by the manufacturer of the spreader, but can be updated by conventional means such as downloading from a disc or form the internet. The reference information can also be automatically modified by the statistics gathered on the spreader. According to one embodiment long term statistics are used to modify reference values of the reference database. If, for example, failure occurs a plurality of times before the parameters or set of parameters monitored indicated that a failure is in the immediate future, the reference value or values are automatically updated. The updated value could be chosen to be a pre-determined percentage lower than the values of the monitored parameters at the moment of failure.

Apart from the reference values and reference curves in the references database, a manufacturer may also provide information on which parameter or sets of parameters need to be monitored to supervise a function. According to a further embodiment of the invention the trend logs are utilised to determine relations between parameters to further refine the prognostic capabilities. It might for example be that the wear of one part under certain conditions, to a high degree can be used to predict the wear of another part. The two parts do not necessarily need to be similar in function or design, but statistic analysis of the trend logs can indicate a statistical relationship between the behaviour of the parts. Thus, the statistics gathered in the trend logs can be used to determine what parameters to monitor for certain functions or parts, and /or the relationship and relative importance of parameters in a set of parameters relating to a function or part.

Methods for this type of statistical analysis are known in the art, for example multivariate analysis and neural network analysis.

The logical view of the diagnostic and prognostic system according to the invention is schematically illustrated in FIG. 6, and the operation will be illustrated by a non- limiting example:

Configuration procedure

The following procedure may be used to set up the diagnostics in the system:

1. choose a diagnostic block (e.g. TWL4).

2. connect input signals to the block.

3. set the parameters for the block where each diagnostic block has it's own set of parameters.

4. connect the block's diagnostic ports to a number of actions (e.g. Create Statistics) .

5. each action is configured with parameters. Each type of action has its own set of parameters (e.g. 'Phone number' and Text' for the action 'Send SMS').

6. each type of action is predefined to use a specific system block (e.g. the action type 'Create Trend' is always using the system block Trend Mgr')

Signals

The signals are the input signals used to feed the diagnostic system with. The signals can be coming from:

- Electrical input and output control signals (digital and analogue) . It includes signals that are electrically located on another device in the system.

- Logical input and output control signals. These values have to be defined by the application engineer within the control program and published to the diagnostic system.

- Diagnostic signals. These signals come from the diagnostic data already stored in the database of the system.

Example 1 : The reference value for the time to move the telescopic arm from position 20' to position 40'.

Example 2: The latest value for the time to move the telescopic arm from position 20' to position 40'.

- System signals. These signals are generated by the system device itself, e.g. the time elapsed since system start and bus errors.

All these signals can be handled in the same way and can be used to diagnose the system.

Diagnostic block A diagnostic block calculates a number of diagnosis based on the input signals. If a certain diagnosis is true the diagnostic block activates the appropriate diagnostic port.

Each diagnostic block is built in advance. A diagnostic block consists of:

- a number of input signals.

- a number of outputs (diagnostic ports). Each diagnostic port represents a specific diagnosis. When a certain diagnosis is set the correct diagnostic port is activated and a certain action is taken, depending on which action block it is connected to.

- a number of parameters that can be used to configure the diagnostic block with.

- the logic that is used to set the correct diagnosis based on the input signals.

User defined connection

The user defined connection is created when an a diagnostic port is connected to a certain action block.

Action block

When a diagnosis is determined to be true a certain action is activated and, if appropriate, the values of the associated parameters are updated and stored in the appropriate database. The action to perform depends on which action block that has been previously selected.

The possible types of actions include:

- Send SMS

- Send Email

- Create Event - Create Alarm

- Create Statistics

- Create Trend

- Stop System

- Start System - Set signal in control system

Each action block has a number of parameters that can be used to configure the action with.

Example 1 : Phone number and Text for the SMS action block.

Example 2: Event number and Text for the Event action block.

System blocks

The system functions are the support functions that encapsulates the functionality to perform the real action, e.g. store data in database or tell the GSM-modem to send an SMS.

The system functions include:

- Messaging Services

- Runtime Mgr

- Event Mgr

- Alarm Mgr

- Statistics Mgr - Trend Mgr

Preferably all the main functions of the spreader are diagnosed mechanically, electrically and hydraulically and, when applicable prognostic, are made according to the above described principles. Parts and functionality suitably include, but are not limited to the following: Twistlocks, Telescope, Flippers, Twin functionality, Twin separating functionality, Gravity point and shift movements.

As an illustrative, but not limiting example the twistlock mechanism will be described

The twistlocks are the lifting mechanism of the Spreader and are usually either operated with hydraulic cylinders or electrical motors A twistlock is schematically illustrated in FIG 5 The twistlocks are designed to turn 90 degrees to alternate between the locked and the unlocked positions The positions of the mechanical elements are monitored by sensors The illustrated twistlock is operated by hydraulic cylinder with two sensors

The following data can be used to determine different issues surrounding the twistlocks, and represent different parameters

The time each movement takes

The pressure required for each movement

The signal level and stability of the sensors monitoring the twistlocks

Example of results/ predictions By reading the trend log to see if the twistlocks are gradually getting slower and comparing to statistics it is possible to determine a recommended time for service prior to mechanical issues leading to a failure

Continuous monitoring of the pressure and comparing this to the pressure drop resulting from the movement, it can be determined if there is a problem with the hydraulic pump or if it is getting harder to move the twistlocks i e mechanical/hydraulic service required

By monitoring the signal stability of the sensors the wear/ aging of the installation can be foretold The signals will gradually flicker at an increasing amplitude (with digital sensors the on/off frequency will increase) as the mechanical adjustment gradually gets worse If not taken care of m time this will lead to an increasing failure rate m the crane directly influencing the MMBF numbers

By having a continuous signal from the sensors it would also be possible to see over-rotation that may not only be cause for premature breakdown, it can also be a safety issue

From the invention thus described, it will be obvious that the invention may be varied in many ways. All such modifications as would be obvious to one skilled in the art are intended for inclusion within the scope of the following claims.