The present invention relates to a method, a computer program product and a system for operating a drilling plant. Some embodiments relate to a system and a method for identifying and/or analyzing aspects of a drilling operation.

BACKGROUND

Identifying and/or analysing the state and operational performance of a drilling operation, for example in petroleum exploration, is challenging due to a number of factors, among others the high number of possible process states (e.g.

drilling, tripping in or out, flow checks, sliding, and bore hole conditioning), the fact that many operational parameters may change quickly, that certain parameters are not directly observable, and the fact that measured signals may contain noise or invalid readings. Although the main activities and their sequence in drilling operations may, in many cases, be fixed, variations regularly occur due to, for example, external disturbances, unexpected equipment failure, manual intervention, or variations in the experience and skills of the drilling crew. Moreover, in many cases, recorded operational data may be erroneous or not representing the actual operating variable accurately (e.g. because of erroneous sensor readings or incorrect input from an operator), but may still be within a range which would be representative for an actual drilling operation. For example, a measured torque and rotation of the drill string to spin in and connect a section of the drill string may be indistinguishable from the start of an actual drilling process, and thereby incorrectly identified as such by a drilling management system.

Some documents which may be useful for understanding the background include US 6,892,812, which describes a method for determining the state of a well operation, wherein measured process data is checked for validity before being used to determine the state, WO 2014/160561 A1 , which describes a method for automatically generating a drilling rig activity report while operating the rig, and US 2015/0167392 A1 , which describes methods for determining the drilling state of a downhole tool and controlling the trajectory of the downhole tool in a wellbore during a drilling operation.

In light of the very high operational cost for drilling plants and drilling rigs, there is a need for improved methods and systems for identifying and analyse operational aspects of a drilling operation, based on measured process parameters and/or manual input from operators. The present invention has the objective to provide an improved method and system, and to obviate at least some disadvantages of prior art techniques.

SUMMARY

Embodiments of the present invention provide a method, a computer program product and a system as set forth in the appended independent claims.

Alternative and/or advantageous embodiments have been set forth in the dependent claims.

BRIEF DESCRIPTION OF THE DRAWINGS

Illustrative embodiments of the present invention will now be described with reference to the appended drawings, in which:

Figure 1 shows an example of sequences in a drilling operation,

Figures 2 and 3 show extracts of the data shown in Fig. 1 and its use in a method for determining the state of a drilling operation,

Figure 4 shows sequences in a drilling operation and their use in a method for determining the state of a drilling operation,

Figure 5 is a schematic flow chart illustrating a method for determining the state of a drilling operation,

Figure 6 illustrates a time sequence with various activities,

Figure 7 illustrates a time sequence with calculated performance indicator parameter values,

Figure 8 illustrates a time-dependent profile for a performance indicator parameter, Figure 9 illustrates a time sequence of performance indicator parameters, Figure 10 illustrates a comparison of performance indicator for different crews, Figure 1 1 illustrates a time sequence of different machine functions,

Figure 12 illustrates a measured and target execution time for a machine function,

Figure 13 illustrates a time sequence of measured execution time for a machine function,

Figure 14 illustrates a time sequence of measured execution time for a machine function for different crews,

Figure 15 illustrates a time sequence of different machine functions, and

Figure 16 illustrates a time sequence of different machine functions.

DETAILED DESCRIPTION During a drilling operation, the drilling plant is operated sequentially, and repeatedly, in a number of different states, such as drilling ("making hole"), connection, reaming, trip in, trip out, etc. Often, these states are referred to as "activities". The activities are largely carried out in a pre-determined sequence during the drilling of a well, however variations may occur. Such variations may be expected, such as the need to replace a drill bit, although their exact timing may not be known in advance, or they may be unexpected, such as an unforeseen need for maintenance or repair of a machine in the drilling plant or other external disturbances. When determining what activity is being carried out, or was carried out at a given time, it is usually assumed that the different activities are separate events, and do not overlap. In that way, one can divide the operation of the drilling rig into a defined set of activities, and find which activity was being carried out at any one time. By logging this operational data, it is possible to evaluate, for example, drilling rig operational efficiency, and identify where performance enhancements may be possible.

An example of various activities which may be defined for this purpose is shown in table 1 . Activity Definition

Drilling The process of making a new hole.

Drilling connection The process of adding a stand of drill pipe to the drill string.

Reaming Return to total depth with pumps on, pipe rotating

and/or weight on bit.

Tripping in The act of running stands into the hole.

Tripping out The act of pulling stands from the hole.

Run casing The time from rigging up to running casing/liner until casing/liner is in position.

Back reaming Pumping and rotating the drill string while

simultaneously pulling out of the hole.

Trip in singles The act of running one single pipe into the hole.

Trip out singles The act of pulling one single pipe out of the hole.

Mud pump ramp-up The act of ramping up the mud pumps to break gel during a drilling connection

Reciprocation The act of cleaning the well during a drilling

connection by axial movement combined with rotation and pumping mud

Friction test The act of measuring friction in the well during a

drilling connection by reading the weight during a hoist - stop and rotate - lower sequence.

Table 1 In addition to this, there may, for example, be defined a general activity encompassing all other, rarer, processes (such as maintenance events), and/or an activity for downtime / idle time.

Such activities or states can be identified manually, i.e. that the drill crew (e.g. the driller) selects which activity, at any time, is being carried out. This can for example be done on a computer-based data logger. This can be a stand-alone unit, or integrated as part of a more comprehensive drilling control system.

Alternatively, the activity can be identified based on measured and logged process variables from the drilling operation. This may be done manually or automatically, for example with a computer-based data analysis system. Figure 1 shows an example of sequences in a drilling operation. Fig. 1 shows a number of logged process variables for the drilling operation, plotted against time. Generally, and to be explained in further detail later, Fig. 1 shows one drilling connection sequence A2, one drilling sequence Ai , and another drilling connection sequence A2.

Figure 1 shows: slips state 1 , where a value 0 indicates slips open and a value 1 indicates slips closed; weight-on-bit (WOB) 2, i.e. the weight acting on the drill bit; rotation 3, i.e. the rotational speed of the drill string; mud flow rate 4; torque 5, i.e. the torque applied to the drill string; and bit depth 6, i.e. the depth of the drill bit in the borehole.

The values for the process variables 1 -6 can be obtained in a number of different ways, e.g. by direct measurement (e.g. sensors), or indirectly by the use of other operational parameters obtained from the drilling equipment. For example, mud flow can be obtained by direct flow measurements, or indirectly via the rotational speed of the mud pumps. Similarly, weight-on-bit can be measured, or calculated from hook load, while taking into account drill string weight. The detailed manner in which these process variables are obtained is not of significance for the present invention.

For the purpose of, among other things, analyzing the drilling operation it is desirable to identify and differentiate between the different activities carried out during the drilling operation, i.e. the different states.

Figures 2 and 3 show extracts of the data shown in Fig. 1 , i.e. a set of historical operational data for a drilling operation over a defined first time interval. This data may be provided from a data logger in a drilling management system. The first time interval may be much longer than what is shown in Fig. 1 , where a relatively short time interval has been used for the purpose of clarity in the description. An activity divider is provided, whereby the weight-on-bit falling below or increasing above a threshold level WOBth will delimit the operational data into a plurality of time intervals. In Fig. 2 it can be seen that weight-on-bit falling below the threshold at time ti to weight-on-bit increasing above the threshold at time t2 provides one time interval ti to t2. Similarly, the time where weight-on-bit again drops below the threshold (see Fig. 3) is a further activity divider, defining a second time interval t2 to tz. Thus, a plurality of second time intervals can be defined for the set of operational data.

For at least one of these plurality of time intervals, at least part of the

operational data falling within that time interval is selected for further analysis. This may include readings for the process variables 1 -6. This step will now be explained in relation to Fig. 3 and the time interval t2 to tz, however the step can be performed equivalently for the time interval ti to t2 and for further time intervals.

Over the time interval, characteristic values of one or more of the process variables may be derived. The characteristic values may, for example, include an average value of the process variable over the time interval, a median value over the time interval, or a change in the process variable over the time interval.

For the process variables shown in Fig. 3, characteristic values may include the average value over the time period for: Rotation, R _avg ; Flow, F _avg ; and Torque, Tavg. They may further include the change in bit depth over the period, ΔΒϋ. The characteristic values can be used to determine the state of the drilling operation for that time interval by comparing the characteristic values with predetermined threshold values. For example, one may define threshold levels /^threshold, ^threshold and Threshold for the rotation, flow and torque levels,

respectively. If the characteristic value for the relevant process variable is above the threshold value for that time period, then one can consider that process variable to be 'true' or 'on'.

By using simple logic or a look-up table, one can then, on the basis of the characteristic values and the threshold(s), determine the state of the drilling operation. For example, to identify a drilling sequence (i.e. "making hole") such as the one shown in Fig. 3 (time interval t2 to tz), the method would identify a time interval starting with an activity divider WOB becoming 'true' (i.e. the logged weight-on-bit increasing above the threshold value) and ending with WOB becoming 'false', and with characteristic values during that time interval indicating that flow F, rotation R, and torque T, are 'on', while the change in bit depth, ΔΒϋ, is above some threshold value, e.g. 5m. This is illustrated in Table 2.

Table 2

Similarly, the time interval shown in Fig. 2 (ti to .2) would be identified as a drilling connection sequence.

The process variable (or variables) used as activity dividers may be included in the operational data. The process variable (or variables) used as activity dividers may also be used to determine characteristic values used to determine the state of the drilling operation. For example, the weight-on-bit may be used as a characteristic value, whereby weight-on-bit is considered to be 'on' if above a pre-determined threshold value. (And this threshold value may, if desirable, be different from the threshold value, WOBth, used for activity divider purposes.) In yet another preferred embodiment, a larger number of defined activities can be used. For this purpose, the method may use more than one process variable as activity dividers. This allows more granularity in the analysis, and thus may give higher accuracy. The number of activity dividers can be determined according to the specific needs in any one case. For example, in addition to weight-on-bit, the slips state can be used as activity dividers. This is illustrated in Fig. 4, in which activity dividers ta-d delimit time periods ta-tb (A3, back reaming: WOB false to slips close), tb-tc (A ₄ , drill pipe connection: slips close to slips open), tc-td (A5, reaming: slips open to WOB true) and td-> (Ai, drilling:

WOB true to WOB false). Thus, by defining several activity dividers it is possible to identify a larger number of different sub-processes during the drilling operation, and allocate these a corresponding activity A _n . This may improve the accuracy and reliability of the analysis. An example of a look-up table for use in such a case is shown in Table 3. The abbreviations used are: bit depth BD; weight on bit WOB; flow F; rotation R; torque T; hole depth HD; stand length SL; and pipe length PL.

Activity dividers Process variable state

Activity Slips

Start End ΔΒϋ WOB F R T

state

WOB= WOB=

Drilling >5 m On On On On 0

True False

Drill pipe Slips Slips

<5 m Off - Off Off connection close open ¹

WOB=

Slips

Reaming True & >10 m Off On On - Open

BD = HD

Slips Slips > SL -

Trip In Off Off Off Off

Close Close 20% ¹

Slips Slips <-SL -

Trip Out Off Off Off Off

Close Close 20% ¹

Back WOB= Slips <-SL -

Off On On - Reaming False Close 20% ¹

Trip In Slips Slips >0 &

Off Off Off Off Singles Close Close <PL ¹

Trip Out Slips Slips <0 &

Off Off Off Off Singles Close Close >-PL ¹

Slips Slips >SL -

Wet Trip In Off On Off Off

Close Close 20% ¹

Wet Trip Slips Slips <-SL -

Off On Off Off Out Close Close 20% ¹

Table 3

An illustrative embodiment of the method for determining a state of the drilli operation has been illustrated by the schematic flow chart in figure 5.

The method starts at the initiating step 1 10. First, the operational data provision step 120 is performed. Step 120 includes providing a first set of operational data of the drilling operation through a first time interval. Typically, the operational data include status data obtained by drilling equipment or operational measurements provided by sensors. The first set of operational data may, e.g., be selected from weight-on-bit, mud

circulation rate, rotation, torque, slips state, and bit depth.

The first set of operational data may, for example, include all of weight-on-bit, mud circulation rate, rotation, torque, slips state, and bit depth.

The provided operational data may include process variables provided directly (e.g. from sensors), or indirectly. For instance, if the operational data includes mud circulation rate, or mud flow, this may be provided either directly by flow measurement equipment or indirectly by monitoring a rotational speed of a mud pump. This has been further elaborated above, e.g. with reference to fig. 1 .

Next, in the activity divider provision step 130, a set of operational activity dividers that delimit a plurality of second time intervals is provided within the first time interval.

The operational activity dividers may advantageously be provided in such a way that the plurality of second time intervals do not overlap. More specifically, the operational activity dividers may advantageously be provided in such a way that the plurality of second time intervals span the entire first time interval.

The operational activity dividers may advantageously represent points of time whereby values of data that are included in the operational data, cross a predetermined threshold value. For instance, an activity divider may represent the point of time whereby a weight-on-bit signal exceeds a predetermined threshold value, or drops below a predetermined threshold value. This has been further elaborated in the above detailed description, e.g. with reference to fig. 2 and 3 above. The operational activity dividers may, e.g., be selected from the following events: slips open, slips close, weight-on-bit on, weight-on-bit off. Other operational activity dividers are also possible. The second time interval may, e.g., be delimited by operational activity dividers that represent data of a same class of operational data. A class of operational data may be a set of operational data that represents the same physical entity. For instance, weight-on bit may represent one class of operational data, while slips state may represent another class of operational data.

Hence, the second time interval may, e.g., be delimited by a first activity divider which represent weight-on-bit on and a second activity divider which represent weight-on-bit off, or vice versa. Alternatively, the second time interval may, e.g., be delimited by a first activity divider which represent slips open and a second activity divider which represent slips closed, or vice versa. In any of these illustrative cases, the same class of operational data is used for delimiting the second time interval. This has been further elaborated in the above detailed description, e.g. with reference to fig. 2 and 3 above. Alternatively, the second time interval may be delimited by operational activity dividers which represent data of different classes of operational data. In such a case, the second time interval may, e.g., be delimited by a first activity divider which represent weight-on-bit off and a second activity divider which represent slips close. In this case different classes of operational data are used for delimiting the second time interval. This principle has been further elaborated and illustrated in the above detailed description, e.g. with reference to fig. 4 and table 2 above.

Referring again to the flow chart of fig. 5, the next step 140 is another

operational data provision step which is performed for at least one of the second time intervals delimited by the activity dividers provided in step 130. Step 140 includes providing a second set of operational data of the drilling operation through the second time interval. The types of operational data provided in step 140 may be a subset of the types of operational data provided in step 120. Alternatively, the operational data provided in step 140 may be the same operational data as those provided in step 120.

As an example, when the first set of operational data includes all of weight-on- bit, mud circulation rate, rotation, torque, slips state, and bit depth, the second set of operational data may include the subset consisting of mud circulation rate, rotation, and torque. Numerous other examples are possible. This has been further elaborated in the above detailed description, e.g. with reference to fig. 3 above.

Further, in step 150, characteristic values of the second set of operational data are derived through the second time interval.

The characteristic values of the second set of operational data through the second time interval may advantageously be calculated as an average of the operational data through the second time interval. Alternatively, the characteristic values of the second set of operational data through the second time interval may be calculated as a median of the operational data through the second time interval, or as a change in the operational data over the second time interval. As an example, average values of mud circulation rate, rotation, and torque may be calculated as characteristic values in step 150. Numerous other examples are possible. This has been further elaborated in the above detailed description, e.g. with reference to fig. 3 above. Next, in step 160, the state of the drilling operation in the second time interval is determined, based on the characteristic values, and the activity can thus be identified. Step 160 of determining the state of the drilling operation based on the characteristic values may advantageously include comparing the characteristic values with predetermined threshold values. For instance, step 160 of determining the state of the drilling operation based on the characteristic values may include looking up in pre-stored data. In this case, the predetermined threshold values may be kept as pre-stored data in a lookup- table, such as table 1 or table 2 referred to in the detailed description above. In one embodiment of the disclosed method, the step 140 of providing a second set of operational data of the drilling operation through the second time interval, the step 150 of deriving characteristic values of the second set of operational data through the second time interval, and the step 160 of determining the state of the drilling operation based on the characteristic values, may be repeated for the plurality of second time intervals that are delimited by activity dividers provided in step 130. This results in a series of determined states of the drilling operation through the first time interval.

In one embodiment of the method for identifying the state of a drilling operation, the step 120 of providing a first set of operational data of the drilling operation through a first time interval is completed before the performance of step 130 of providing a set of operational activity dividers, step 140 of providing a second set of operational measurements, step 150 of deriving characteristic values, and the step 160 of determining the state of the drilling operation. This allows the state of the drilling operation and the activity for a given time interval to be determined in a post-processing analysis, after the completion of the

operational data acquisition in step 120.

The illustrated method ends at the terminating step 190.

Now referring to Figs 6-16, embodiments of a method for analyzing a drilling operation will be described.

The method comprises providing a first set of logged operational data from the drilling operation through a first time interval ; providing a set of operational activity dividers that delimit a plurality of second time intervals t _n within the first time interval; and allocating one of a plurality of pre-defined activities A _n to each of the second time intervals.

The operational activity dividers may be provided from an operator, may be received from a drilling control system, may be calculated according to the method described above and in relation to Figs 1 -5, or may be calculated from operational data according to a different method.

Fig. 6 illustrates a time sequence having eight time intervals ti-ts, each having been allocated an activity A _n . In this illustrative and simplified example, alternating drilling (Ai) and connection (A2) sequences are shown, only with one intervening, different activity Αβ at time interval te. The activity Αβ may be, for example, an unexpected disturbance, or planned maintenance which needs to be carried out on the drilling plant. A problem with such intervening activities is that they will influence the overall performance of the drilling plant, for example the number of meters drilled per hour, however a deterioration of the

performance due to such activities is not indicative of a fundamental reduction in performance of the drilling plant itself (such as the drilling machine or drill floor equipment) or the performance of the drilling crew. Such unexpected or "intervening" activities therefore complicate any calculation and quantification of drilling plant and/or crew performance. It will be appreciated that the time period considered, i.e. the first time interval, may be considerably longer and comprise a significantly higher number of activities than that which is, for clarity, illustrated in a simplified manner in Fig. 6.

According to one embodiment, the method comprises the step of, for at least one second time interval having been allocated a first activity, the first activity being one of the plurality of pre-defined activities, calculating a first performance indicator parameter, the first performance indicator parameter being calculated as a function of the duration of the respective second time interval having been allocated the first activity. Fig. 7 illustrates a first performance indicator parameter di-d ₄ having been calculated for the sequence illustrated in Fig. 6. In Fig. 7, the first performance indicator parameter, f(t), is simply equal to the duration of each time interval ti-ts having been allocated the relevant activity, hence the first performance indicator parameter di-d ₄ is indicative of the duration of the drilling activities Ai.

Alternatively, the first performance indicator parameter may, for example, be a function of the duration of each time interval ti-ts and a pre-determined, target time duration of the respective activity. This could be, for example, include quantifying the absolute or relative difference in the measured time and the target time for that activity and determining the performance indicator parameter based on this. (Hence, a measured time being larger than the target time could lead to a poorer value for the performance indicator.) Fig. 7 further shows a calculated second performance indicator parameter e-i-e3, which is indicative of the duration of the connection sequences A2.

While the value f(t) of the first and second performance indicator parameter in Fig. 7 are illustrated graphically in a column diagram, it is clear that any method of logging and/or displaying these values may be employed.

In the example illustrated in Figs 6 and 7, values for the performance indicators are calculated for a plurality of second time intervals ti-ts. Having values for several time intervals allows the performance indicator to be compared at different points in time, whereby a change, such as a deterioration of the performance, can be identified. For example, if the duration of the connection sequences A2 increase over time, that may be an indication of deteriorating performance of some drill floor equipment. Such identification of deteriorating performance may be done on the basis of one (first) performance indicator parameter di-d ₄ , and comparing calculated first performance indicator parameter values at different points in time. Alternatively, or additionally, one may use the second performance indicator ei-e3 for such purposes. Since many machines in a drilling plant carry out operations in several different activities, this provides more accuracy, in that one can correlate the indications provided by each performance indicator parameter. For example, if a deteriorating performance of a machine is indicated by the first performance indicator parameter but not by the second, and where it would have been expected to see a similar pattern (deteriorating performance) in the second performance indicator parameter, that may be taken as an indicator that the problem lies elsewhere, for example in the operation by the crew. In this way, higher accuracy is achieved when evaluating performance according to the method disclosed here.

In an embodiment, a time-dependent profile for the performance indicator is created for the plurality of time intervals. Fig. 8 illustrates an example of this, showing the development of one of the performance indicators (in this case performance indicator d _n ) over time. If, as illustrated, a deteriorating trend (indicated by the arrow) in the performance indicator over time is identified, further investigations, corrective maintenance, or replacements/repairs may be carried out (or planned) to ensure that the overall performance of the drilling plant remains satisfactory. Alternatively, or additionally, the performance indicator value, or values, may be compared to a pre-determined target performance indicator value. This may, for example, be a threshold value which is indicative of a need for maintenance, replacement or repair of a machine in the drilling plant. The method may further comprise the steps of performing a maintenance activity on a machine associated with a drilling plant carrying out the drilling operation, replacing a component of the machine, or replacing the machine.

In one embodiment, the performance indicator may be calculated as a function of the duration of a pre-determined set of several second time intervals. This may, for example, be the time intervals lying between two particular events, such as any activity occurring between slips close and the next slips close. Consequently, one can obtain a measure of the performance between these events, even if multiple, different activities are carried out. Referring to Fig. 6, such a performance indicator being a function of several time intervals may include ti and t2 in order to get an indication of the

performance between "WOB on" and the next "WOB on". (WOB on indicating the start of a drilling activity Ai , and this period thus being indicative of the time between each stand.) Additional values for the performance indicator may be calculated for the time periods t3 and U, as well as ts, te and t ₇ , and so on.

Alternatively, the performance indicator may be a function of the time (i.e.

duration) between two operational activity dividers. Consequently, one could use the time between activity dividers directly to calculate the performance indicator. For example, with reference to Fig. 4, this may be the time between WOB off (t _a ) and WOB on (td). Alternatively, it might be beneficial to determine performance indicators between other activity dividers, such as WOB off (t _a ) and slips on (tb). This may, for example, allow more accurate performance evaluations of specific machines, which carry out their main machine functions during this time interval and/or there is less potential disturbance and noise from other machine functions during that time interval. Fig. 9 illustrates part of a real data set from a drilling operation, comprising similar information as in Figs 6 and 7, where performance indicators for in slips time, open slips time and slips to slips time are calculated.

Having at least two values for the performance indicator (or indicators) according to any of the options described above, one can compare values for the performance indicator obtained at different times during the drilling operation. This allows a comparison of the operational performance at different times. It may, for example, be beneficial to compare the performance between different drilling crews to identify possible improvement potentials, the need for (or effect of) crew training, or the like. This is illustrated in Fig. 10, with a performance indicator ΪΑ,Β is illustrated for different crews A and B. The performance indicator ΪΑ,Β may, in this case, for example be the average in slips time (see Fig. 9) over a sample period of operation. The information that Crew B operates more efficiently (in this example, completes the in slips activities on average in a shorter time), may be used to target training efforts for those particular operations to Crew A. By means of the method according to this embodiment, it is therefore possible to identify such differences and

improvement potentials with more granularity, more accuracy and less risk of the findings being influenced by external events (for example, unexpected interruptions such as that exemplified as Αβ in Fig. 6) or other factors which are not related to the drill crew performance.

Additionally, or alternatively, one may compare the effects of external factors occurring at different times, such as weather conditions or events in the well or reservoir, in an objective manner, i.e. study their influence individually on specific activities. This may allow for a better understanding of the different operations, which may be used for improved operational planning and/or improved system design.

Typical performance indicator parameters which may be useful to identify and base an analysis upon include: in slips time; open slips time; slips to slips time; time between two drilled stands; weight to slips time; and slips to weight time. Several, or a combination, of these may also be used in the analysis.

A method for analyzing a drilling operation according to any of the above embodiments thus permits a more accurate and detailed analysis of the operation. Identifying the activity dividers and allocating activities to the different time intervals may be done in any manner, for example these may be received from an operator during the operations. For this purpose, an operator may have an interface (such as a touchscreen panel) which allows the operator to at any time select what activity is being carried out, and where this selection would be stored in a time series. Alternatively, the activity and/or activity dividers may be calculated automatically by a drilling control system, and stored. In another alternative, the activities may be determined with a method as described above in relation to Figs. 1 -5.

In one embodiment, there is provided a method for analyzing a drilling operation, comprising the steps: (a) obtaining a first set of logged operational data from the drilling operation through a first time interval;

(b) obtaining a set of operational activity dividers that delimit a plurality of second time intervals t _n within the first time interval;

(c) allocating one of a plurality of pre-defined activities A _n to each of the second time intervals, and

(d) for at least one second time interval, identifying a plurality of machine functions carried out during the second time interval and a measured execution time for each of the machine functions.

Steps a-c can be performed similarly as described above. Allocating activities to each of the second time intervals may be carried out similarly as described in relation to Figs 1 -5 above, or in a different manner. Fig. 1 1 illustrates in a simplified manner a plurality of machine functions carried out during two activity periods, Ai (drilling) and A2 (connection). The machine functions include individual actions or sub-processes carried out by the various machines in the drilling plant, during some activity. The illustrated machine functions in Fig. 1 1 have, for convenience, been labelled a-t. The different machines are denoted by their abbreviations, where DDM is derrick drilling machine, DW is drawworks, VPH is vertical pipe handling machine, HRN is hydraulic roughneck, and RT is rotary table. Further, FB denotes fingerboard, WC denotes well centre, and LGA denotes lower guide arm. The start time, end time, and execution time for each machine function is shown by the horizontal columns, which are drawn against a time axis t.

As will be clear, a number of other machine functions will also be carried out during, for example, a connection sequence. A selection of such machine functions, as illustrated in Fig. 1 1 , can be used in the analysis. The selection can include a larger number of machine functions than that illustrated in Fig. 1 1 , in order to increase granularity and/or accuracy of the analysis. A lower number may also be used, if, for any given requirements, that is sufficient for the desired purpose and analysis. Fig. 1 1 shows the start time, the end time and the duration of each machine function. These parameters can be identified graphically from a plot such as that shown in Fig. 1 1 , and/or they can be calculated by a drilling control system and displayed to an operator.

The method can also include receiving a target execution time for each machine function. The target execution time can be compared with the measured execution time in order to identify any deviation. In Fig. 1 1 , a measured execution time which exceeds the target execution time is shown in solid filled sections of each column. As can be seen, some machine functions, such as machine function n. VPH move towards FB, exceed their target execution time. By comparing and displaying the difference between measured execution time and target execution time, one may thereby identify an underperforming machine function. Such underperformance may be a result of, for example, increased wear, damage or some malfunction on the machine, or it may be the result of inexperienced crew and/or non-optimized operation of the machine.

Fig. 12 shows an alternative presentation of this information, with the measured and target execution time for machine function b. DDM dolly retract.

The analysis can be repeated for a plurality of second time intervals. This may be a plurality of time intervals having the same activity, for example A2 drilling connection. One can thereby compare the measured execution time for one or more of the machine functions over a number of intervals. For example, one may compare the execution time of one machine function over a series of time intervals having the same activity allocated to them, i.e. the same machine function performing the same type of operation. In this way, one can compare the performance of the machine function under comparable operating conditions, thus achieving more accuracy and a more reliable and useful result.

Figure 13 shows an example of the development in the measured execution time T (see Fig. 12) of machine function b. DDM dolly retract. A increasing trend can be identified, indicating that the performance of the machine is deteriorating over time. Fig. 14 shows a different example, in which a difference between two crews can be analyzed. In this case, one may identify that Crew B may benefit from more training or a review of the working procedures, in order to obtain the same efficiency as Crew A. By means of the method according to embodiments disclosed herein, such analyses can thus be carried out and done with more accuracy and/or with more granularity.

In one embodiment, the method comprises, for at least one of the plurality of machine functions, identifying a difference between a target execution time and a measured execution time, and identifying a delay in an operational activity divider caused by the difference. An example is illustrated in Fig. 15. In this case, when the connection sequence A2 has completed, machine function a. DDM drilling will commence, and the operational activity divider slips open will cause the identified activity to change to Ai drilling. A prolonged execution time for a machine function during the preceding connection sequence A2 may, however, delay the start of drilling, if the start of drilling is dependent on that machine function having completed. For example, in Fig. 15, a prolonged execution time x (indicated by the solid filled area of the column, being the difference between the target execution time (hatched) and the actual, measured execution time) for the machine function s. RT slips open leads to a delay in the operational activity divider 'slips open', and thus to an increased duration of the connection sequence A2 and a delay in the start of drilling Ai.

By means of this embodiment, it is possible to identify the influence of such a prolonged execution time in a machine function on the overall drilling plant operation, i.e. the extent to which, for example, a deteriorating machine function performance actually influences the overall operation, or whether it has a negligible or no influence. As can be seen from Fig. 15, a prolonged execution time for certain other machine functions do not influence overall operations. This includes, for example, machine function n. VPH move towards FB and r. HRN elevate down, as these are machine functions which can be performed (or completed) in parallel with the drilling Ai commencing. However, this is not the case for machine function s. RT slips open.

Another example is illustrated in Fig. 16, in which a prolonged execution time y in the machine function p. HRN lower clamp open leads to a delay for the subsequent, dependent machine functions q-s. This includes machine function s. RT slips open, which must be carried out subsequent to p. HRN lower clamp open and prior to the start of drilling (machine function a. DDM drilling). This delay therefore has a direct influence on drilling plant operation, unlike, for example, a prolonged execution time in machine function n. VPH move towards FB. According to this embodiment, it is therefore possible to identify such performance reductions and identify their influence on the overall plant operation. One can thereby identify machine functions which underperform, which deteriorate over time and quantify the effects of this, or machine functions which are not carried out in an optimal manner by the crew, whereby additional training might improve performance. Additionally, or alternatively, one can use the method according to this embodiment to study the effects of replacing a machine with an improved machine having lower target execution times, thereby possibly achieving performance enhancements for those machine functions carried out by that machine. The effects of this on the overall drilling plant operation may thereby be identified and quantified.

Using methods according to the above embodiments to analyze a drilling operation, the activity for given time intervals, and performance indicators for activities and/or machine functions permits an accurate evaluation of the performance of the drilling operation, with less sensitivity to e.g. erroneous or noisy data, natural variations (e.g. torsional vibrations in the drill string producing short torque peaks), or human error or variations. The latter may be the case, for example, if the analysis is based on data entered by the operator him-/herself. Different persons may enter, for example, a change in activity at different times, and such data may be subject to external disturbances which may influence the logged data but are not actually representative of a problem with the machines or the crew. Methods according to the present invention are less sensitive to such variations, and thus determine the performance of a drilling operation and relevant sub-processes and operations more accurately, and more consistently and comparable over longer time periods.

For example, in accordance with some embodiments described above, the time intervals are allocated different activities on the basis of the actual drilling operation, i.e. according to what the drilling plant actually carried out. This may be different from what is commanded by the driller, as there may be time delays between a command is sent to carry out some operation, and until the operation actually starts. In order to determine performance parameters accurately (e.g. the execution time for different machine functions) and to compare these over time, it is beneficial to base such calculations on comparable activities. For example, the performance of various machine functions may be different during trip in singles compared to tripping in of regular drilling stands. Hence, a direct comparison of, for example, execution time for machine functions in drill floor equipment, may suggest a deteriorating performance which is, in fact, not real. By allocating activities with sufficient granularity, such effects can be reduced or eliminated, and the true and actual trends of deteriorating performance for a machine function may be determined more accurately.

Methods as described above may advantageously be implemented as a computer-implemented method. For such a case, a computer program product has been provided, which when loaded into a memory and executed on a processing device causes the processing device to perform a method as described above.

Also, the method may be implemented in a system for analyzing a drilling operation. Such a system comprises input devices for providing operational data of the drilling operation; and a computer device, configured to perform a method as disclosed herein. More specifically, the computer device may include a processing device and a memory, the memory being arranged to hold a computer program that causes the processing device to perform a method as disclosed herein when the computer program is executed by the processing device.

The invention is not limited to the embodiments described herein; reference should be had to the appended claims.