SE541415C2

SE541415C2 - Method and system for prediction of a drainage valve malfunction probability

Info

Publication number: SE541415C2
Application number: SE1651266A
Authority: SE
Inventors: Jonas Biteus; Martin Eineborg; Samia Khalid; Tony Lindgren
Original assignee: Scania Cv Ab
Priority date: 2016-09-26
Filing date: 2016-09-26
Publication date: 2019-09-24
Also published as: BR102017020119A2; SE1651266A1; DE102017008611A1

Abstract

A method and a system for prediction of a probability for a malfunction of a drainage valve of an air dryer included in a pressurized air system in a vehicle are presented. The drainage valve is arranged for discharging humidity being absorbed from pressurized air flowing through the air dryer. According to the present invention, the prediction of the probability for a malfunction of the drainage valve indicates the probability for the malfunction during an upcoming prediction time period tof predefined length. Also, the prediction is determined based at least on one or more parameters being related to a use of the vehicle.

Description

METHOD AND SYSTEM FOR PREDICTION OF A DRAINAGE VALVE MALFUNCTION PROBABILITY Field of invention The present invention relates to a method for prediction of a probability for a malfunction of a drainage valve of an air dryer. The present invention also relates to a system arranged for prediction of a probability for a malfunction of a drainage valve of an air dryer. The present invention also relates to a computer program and a computer program product implementing the method according to the invention.

Background of invention The following background information is a description of the background of the present invention, which thus not necessarily has to be a description of prior art.

Vehicles of today often include a pressurized air system.

Pressurized air is used by one or more pressurized air consumers in the vehicle, and is used e.g. for controlling and/or driving many devices and/or systems in vehicles, such as e.g. different kinds of brake systems, air suspension systems, door control systems, differential gear systems and/or accessory/additional systems.

The pressure of the air is increased/created by use of a compressor often being driven by the engine. The pressurized air may then include some moisture/humidity and/or oil impurities. This moisture/humidity in the air may cause condensation into water in piping, tubing, or valves providing the air to the consumers, and/or may cause condensation within the air consumers. This water may in its turn cause problems in systems being controlled and/or driven by the pressurized air. For example, water located within an air suspension system may impair the suspension function. Also, water may freeze at low temperatures, and may then cause plugging of pipes and/or tubes in the vehicle, which may result in malfunction of important vehicle systems. Further, metal parts in the pressurized air system, such as e.g. in pipes, valves and/or in one or more air consumers, may become rusty due to the moisture/humidity in the pressurized air. Such rust may impair the function of these parts, and may cause malfunctions and/or early exchanges of one or more parts.

Therefore, in order to reduce problems related to the moisture/humidity in the pressurized air, the pressurized air is led through an air dryer included in the pressurized air system. In the air dryer, humidity/moisture in the pressurized air is absorbed and is discharged through a drainage valve of the air dryer, whereby the pressurized air is dried. Hereby, dry pressurized air may be provided to the consumers of pressurized air in the vehicle.

SUMMARY OF INVENTION The function of the drainage valve of the air dryer is important for the air drying capability of the air dryer. If the drainage valve does not open properly to discharge the absorbed humidity/moisture from the air dryer, the pressurized air will not be sufficiently dried in the air dryer.

Therefore, additional drainage sensors have, according to some prior art solutions, been applied on the drainage valve in order to monitor the function of the drainage valve. However, such additional drainage valve sensors add to the hardware complexity, and thus also to the manufacturing costs, for the vehicle. Also, when the additional drainage sensor indicates that there is a problem with the drainage valve, it might be too late to avoid that the vehicle must be taken off road (Vehicle Off Road) in order to take care of the drainage valve problems .

According to some other prior art solutions, the regular maintenance is increased, i.e. the maintenance intervals for the air dryer and the drainage valve are shortened in order to prevent drainage valve malfunction. However, to increase the maintenance obviously increases the total cost for the vehicle maintenance, since well-functioning drainage valves will then often unnecessarily be serviced and/or exchanged. Also, to shorten the maintenance intervals forces the vehicle to spend more time in the garage, which is not very attractive for a vehicle owner.

It is therefore an object to solve at least some of the above mentioned disadvantages.

The object is achieved by the above mentioned method for prediction of a probability for a malfunction of a drainage valve of an air dryer included in a pressurized air system in a vehicle, wherein the drainage valve is arranged for discharging humidity being absorbed from pressurized air flowing through the air dryer. According to the present invention, the prediction indicates the probability for the malfunction during an upcoming prediction time period tpredof predefined length. Also, the prediction is determined based at least on one or more parameters being related to a use of the vehicle.

According to an embodiment of the present invention, the one or more parameters include a parameter related to a time period ttemPduring which the vehicle including the drainage valve has been in operation at an ambient temperature Tambientwithin a predefined temperature interval Tambient_intervai· According to an embodiment of the present invention, the one or more parameters include a parameter related to a distance D travelled by the vehicle including the drainage valve.

According to an embodiment of the present invention, the one or more parameters include a parameter related to at least one time period tpressure_1,tpressure_2 , ...tpressure_nduring which the vehicle has been in operation within at least one ambient pressure interval Pambient_interval_l, Pambient_interval_2,...Pambient_interval_nrespectively .

According to an embodiment of the present invention, the one or more parameters include a parameter related to a type of the vehicle including the drainage valve According to an embodiment of the present invention, the one or more parameters include a parameter related to a time period tveiocityduring which the vehicle including the drainage valve has travelled in a predetermined velocity interval interval· According to an embodiment of the present invention, the prediction is also based on a time period tregenerationduring which the air dryer has been in a regeneration phase.

According to an embodiment of the present invention, the prediction is also based on a time period theatingduring which the air dryer has been in a heating phase.

According to an embodiment of the present invention, the prediction is determined using one or more of: - a decision tree algorithm; and - a generalized boosted regression model (GBM) algorithm.

According to an embodiment of the present invention, the upcoming prediction time period tpredis 15 days.

The object is also achieved by the above mentioned system arranged for prediction of a probability for a malfunction of a drainage valve of an air dryer. The system includes a control unit arranged for determining the prediction such that it indicates the probability for the malfunction during an upcoming prediction time period tpredof predefined length, and for determining the prediction based at least on one or more parameters being related to a use of the vehicle.

According to an embodiment of the present invention, the control unit is arranged to include in the one or more parameters a parameter related to a time period ttemPduring which the vehicle including the drainage valve has been in operation at an ambient temperature Tambientwithin a predefined temperature interval Tambient_interval.

According to an embodiment of the present invention, the control unit is arranged to include in the one or more parameters a parameter related to a distance D travelled by the vehicle including the drainage valve.

According to an embodiment of the present invention, the control unit is arranged to include in the one or more parameters a parameter related to at least one time period tpressure_1 ,tpressure_2, ...tpressure_nduring which the vehicle has been in operation within at least one ambient pressure interval Pambient_interval_1, Pambient_interval_2,... Pambient_interval_nrespectively .

According to an embodiment of the present invention, the control unit is arranged to include in the one or more parameters a parameter related to a type of the vehicle including the drainage valve According to an embodiment of the present invention, the control unit is arranged to include in the one or more parameters a parameter related to a time period tvelocityduring which the vehicle including the drainage valve has travelled in a predetermined velocity interval vinterval · According to an embodiment of the present invention, the control unit is arranged to base the prediction also on a time period tregenerationduring which the air dryer has been in a regeneration phase.

According to an embodiment of the present invention, the control unit is arranged to base the prediction also on a time period theatingduring which the air dryer has been in a heating phase .

According to an embodiment of the present invention, the control unit is arranged to determine the prediction by use of one or more of: - a decision tree algorithm; and - a generalized boosted regression model (GBM) algorithm.

The object is also achieved by the above mentioned computer program and computer program product.

The present invention provides for a simple and reliable avoidance of the above mentioned problems related to humid pressurized air being provided to parts of the pressurized air system and/or to the air consumers. Since the air drying function of the air dryer may be predicted in advance according to the present invention, actions may be taken well in time to avoid that humidity related problems occur. Hereby, the risk for plugging, rust and/or malfunction of air consumer systems is considerably decreased, since the air dryer may be serviced already before the problems start occurring.

By usage of the present invention, a conditioned based maintenance of the air dryer, and more specifically of the drainage valve, is provided. Drainage valves having an increased risk for having a malfunction may be identified in advance by the prediction provided by the present invention. Hereby, vehicles including such drainage valves being, or going to be, in the risk zone may be identified and may be requested to be serviced. This makes it possible to provide a vehicle individual adaptive maintenance of the air dryer and/or drainage valve, which secures that proper maintenance is provided in order to avoid vehicle off road situations, at the same time as the total maintenance cost for the vehicle is lowered .

The prediction of the malfunction probability according to the present invention may be based on information normally already being available in a vehicle, whereby a minimal addition to the vehicle hardware complexity arises from implementation of the prediction according to the present invention. For example, the predictions may here be based on sensor signals and or control signals normally being provided and used in vehicles of today, such as sensor signals from an ambient temperature sensor, an ambient pressure sensor and/or a vehicle velocity sensor. Also, the predictions may be performed by an external control/computation unit located outside of the vehicle, which also minimizes the added software complexity of the vehicle.

The exactness/reliability of the drainage valve malfunction probability prediction may be adapted according to an embodiment of the present invention. This adaption may be performed by an intelligent selection of the parameters, and also of the number of parameters, on which the prediction is based. Since the exactness/reliability of the prediction may be adapted, also the computational complexity may be adapted, since there is a relation between the exactness/reliability and the computational complexity.

Detailed exemplary embodiments and advantages of the method, system, computer program and computer program product according to the invention will now be described with reference to the appended drawings illustrating some preferred embodiments .

BRIEF DESCRIPTION OF THE DRAWINGS Embodiments of the invention are described in more detail with reference to attached drawings illustrating examples of embodiments of the invention in which: Figure 1 is a schematic illustration of a vehicle which may include a system according to the present invention, Figure 2 is a schematic illustration of a drainage valve, Figure 3 is a flow chart diagram for a method according to an embodiment of the present invention, Figure 4 is a schematic illustration of a control unit.

DETAILED DESCRIPTION OF INVENTION Figure 1 schematically illustrates an engine 101 and some parts of an example system 150 arranged for providing pressurized air in a vehicle 100. In figure 1, a system 150 providing pressurized air to an air consumer being exemplified as a brake system is illustrated. However, the pressurized system 150 may also be arranged for providing pressurized air to one or more other air consumers in the vehicle, such as for example other air consumers mentioned in this document. Also, as is understood by a skilled person, a pressurized air system 150 may include more parts than shown in figure 1. Figure 1 is simplified in order to increase intelligibility, and is mainly used for explaining the basic principles of the present invention in relation to a pressurized air system 150.

The pressurized air system 150 includes a compressor 102, which may be driven by the engine 101 of the vehicle 100. The compressor 102 outputs pressurized air 131 to a pressure regulator 103. The regulator 103 outputs pressurized air being input 132 to an air dryer 104. The pressurized input air 132 being input to the dryer 104 may, as is mentioned above include moisture/humidity, which should be removed from the air flow. In the air dryer 104, the moisture/humidity is separated from the pressurized air, as is described more in detail below. Thus, the pressurized air 133 being output from the air dryer 104 is essentially dry, i.e. is essentially free from moisture/humidity. This essentially dry pressurized air 133 may then be provided to one or more tanks 105, 106 arranged for storage of pressurized air.

In the non-limiting example in figure 1, the dried pressurized air is in pipes and/or tubes 107, 108 provided to brake systems 121, 122, 123, 124 arranged at the wheels 111, 112, 113, 114 of the vehicle 100. These brake systems 121, 122, 123, 124 utilize the pressurized air to control the applied braking force and/or provide the actual braking force. As is known for a skilled person, compressed air may be used in a number of ways in different brake systems, such as e.g. in anti-lock/skid braking systems.

The pressurized air system 150 may be controlled by an air system control unit 160. The air system control unit 160 is in figure 1 illustrated as being connected to the compressor 102 and to the air dryer 104. However, the air system control unit 160 may also be connected to one or more of the other parts of the pressurized air system 150 although not illustrated in figure 1. The air system control unit 160 may also be connected to other control units 190 in the vehicle 100, to one or more devices in the vehicle and/or to one or more sensors 191 in the vehicle 100, as described below. The air system control unit 160 may be connected to a prediction and/or communication control unit 170, as is explained more in detail below.

Figure 2 is a simplified schematic illustration of some parts of the air dryer 104, which may be used for explaining the principle of the air dryer 104, and the drainage valve 230.

Pressurized air 132 possibly including moisture/humidity is input to the air dryer 104 via an inlet 241. The air flow 230 is led through a receptacle of the dryer 104 and is output 133 through an outlet 242. Within the receptacle, a number of small-sized balls/pellets 210 are arranged such that the air flow 230 comes in contact with the balls/pellets. The balls/pellets then absorb the moisture/humidity of the air flow 230. The moisture/humidity is condensed, and is by force of gravity led 220 to the bottom of the dryer 104 as water. A drainage valve 230 is arranged at the bottom of the dryer 104 for passing water 243 out of the dryer 104. The opening/closing of the drainage valve 230 may be related to the pressure regulator e.g. such that it is opened when the pressure reaches a high/maximum value. Hereby, the water/moisture/humidity absorbed in the air dryer 104 may be discharged 243 through the drainage valve 230, whereby the output air flow 133 is dried from moisture/humidity/water. For some implementations, a drainage valve heater 250 is arranged at the drainage valve 230. In order to reduce the risk that the water 243 freezes in the drainage valve 230, thereby causing plugging of the drainage valve 230 and possibly also a malfunction of the dryer 104, the drainage valve heater 250 may enter a heating phase, which increases the temperature of the valve 230.

An air dryer 104 may be configured in a large number of ways, including e.g. one or more receptacles of suitable sizes, one or more sorts of balls/pellets of suitable sizes and/or a suitable number of balls/pellets.

The air dryer may be provided with an air dryer control unit 270, which may be arranged internally in the air dryer 104 and/or externally of the air dryer 104. The air dryer control unit 270 may be arranged for controlling the function of the air dryer 104, for controlling the drainage valve heater 250 and/or for controlling a regeneration of the air dryer. During a regeneration phase of the air dryer, water being absorbed in the air dryer 104 is emptied, i.e. the air dryer is regenerated. The air dryer control unit 270 may be connected to the air system control unit 160 mentioned above.

Figure 3 shows a flow chart diagram for a method according to an embodiment of the present invention, i.e. of a method for predicting a probability for a malfunction of a drainage valve 230 of an air dryer 104 included in a pressurized air system 150 in a vehicle.

In a first step 301 of the method, one or more parameters related to a use of the vehicle including the air dryer 104 are determined. These parameters may for example be provided by one or more sensors, by one or more control/calculation units and/or by one or more models, as is described more in detail below. Generally, parameter values may in a vehicle be based on measurements, e.g. performed by sensors, and/or may be based on models, such as prediction models. A large number of vehicle parameters are normally already used in vehicle systems of today. Essentially, all such parameters may be used by the method of the present invention.

In a second step 302 of the method, the prediction of the probability for a malfunction of the drainage valve 230 during an upcoming prediction time period tpredof predefined length is determined/calculated. The prediction is here determined/calculated based at least on the one or more parameters determined in the first step 301 of the method. As described below, the prediction may also be based on other parameters .

In a third step 303 of the method, the prediction indicating the malfunction probability determined in the second step 302 it utilized. For example, the prediction may be presented to a driver in a driver interface, to a fleet manager and/or to an owner of the vehicle via a suitable messaging/information system. Based on this presented prediction, the driver, the fleet manager and/or the owner may take appropriate actions to avoid vehicle problems and/or standstill, e.g. take the vehicle to a repair shop.

The upcoming prediction time period tpredmay have a length long enough for the driver/manager/owner of the vehicle to have a possibility to service the drainage valve 230, i.e. to take the vehicle to a vehicle repair shop or the like, before the malfunction of the drainage valve actually occurs. The upcoming prediction time period tpredmay also be short enough to not annoy the driver/manager/owner with warnings too early. According to a non-limiting example, the prediction time period tpredmay have a length corresponding to 15 days.

Hereby, actions may be taken in time to avoid that humidity related problems, such as plugging, rust and/or malfunction of air consumer systems, occur due to a malfunctioning drainage valve. A conditioned based maintenance of the air dryer is thus provided, which reduces the total maintenance cost for the vehicle.

Also, the prediction time period tpredmay, when the present invention is used, be intelligently chosen, e.g. such that it ends essentially when a next already planned vehicle service is to take place. Hereby, the prediction indicating if a malfunction within the prediction time period tpredis probable also becomes a prediction of if it is necessary to book an extra service before the already planned service or not.

The one or more parameters related to the use of the vehicle may include a suitable number of parameters. Different parameters add different contributions to a prediction reliability/exactness. Often, the more parameters the prediction is based on, the more reliable the prediction becomes. Therefore, by carefully choosing which parameters and/or the number parameters to be used as a basis for the prediction of the drainage valve malfunction probability, the reliability/exactness of the prediction may be adapted.

According to an embodiment of the present invention, the specific parameters and/or the number of parameters to be included in the prediction may be adapted based on characteristics of the drainage valve, such as the manufacturer, and/or other vehicle specifications, such as an engine size, a power take off, which axles/shafts being provided with air suspensions and/or a type of the vehicle, e.g. a truck, a timber vehicle, a van, a long distance bus or a city bus.

According to an embodiment of the present invention, the one or more parameters related to the vehicle use include a parameter related to a time period ttempduring which the vehicle 100 including the drainage valve 230 has been in operation at an ambient/surrounding temperature Tambientwithin a predefined temperature interval Tambient_interval ·The ambient temperature Tambientmay here be measured e.g. by a standard ambient temperature sensor 191 in the vehicle. The time period ttempis here incremented/measured when the ambient temperature Tambientis within the predefined temperature interval Tambient_interval ·According to an embodiment of the present invention, the predefined ambient temperature interval Tambient_intervalis delimited by -20°C and -10°C, whereby the time period ttemPis incremented by an ambient temperature within the interval; -20°C < Tambient< -10°C. The time period ttemPis reset if the air dryer 104 and/or the drainage valve 230 are exchanged and/or serviced.

According to an embodiment of the present invention, the one or more parameters being related to vehicle use, and on which the prediction is based, include a parameter related to a distance D travelled by the vehicle 100 including the drainage valve 230. This parameter may e.g. be provided by a standard vehicle speed sensor in the vehicle 100 and/or by another suitable device being able to determine distances D covered by the vehicle, such as a navigation device possibly including at least one positioning sensor, e.g. a Global Positioning System (GPS) sensor. The distance D may be compared to a distance threshold Dthresholdof a group of one or more distance thresholds and/or to a distance interval Dintervalof a group of one or more distance intervals to see if there is a risk for a malfunction of the drainage valve 230. These distance thresholds and/or distance intervals may have values chosen depending on at which stage of the prediction algorithm, e.g. at which stage of a used decision tree algorithm and/or a used generalized boosted regression model algorithm, the distance thresholds and/or distance intervals are to be utilized.

According to an embodiment of the present invention, the drainage valve malfunction prediction is also based on a time period tregenerationduring which the air dryer 104 has been in a regeneration phase. The regeneration phase time period tregenerationmay be determined based on regeneration phase control signals being provided by the air dryer control unit 270. The air dryer 104 is used for drying the pressurized air in the pressurized air system, as is explained above. After some time of usage, the air dryer 104 includes some absorbed water and needs to be emptied of water, i.e. needs to be regenerated, in order to function well as a dryer. The air dryer 104 is regenerated by blowing pressurized air through the air dryer and the drainage valve 230, e.g. in a direction from the outlet 242, passing by the balls/pellets 210, and out through the drainage valve 230. Hereby, water/moisture within the dryer 104 is carried/blown away 243 from the air dryer 104, and its air drying capability is increased again. The time period tregenerationmay be compared to a time threshold tregeneration_thresholdof a group of one or more time thresholds and/or to a time interval tregeneration_intervalof a group of time intervals. These time thresholds and/or time intervals may have values chosen depending on at which stage of the prediction algorithm the time thresholds and/or time intervals are to be utilized.

According to an embodiment of the present invention, the prediction of the drainage valve malfunction probability is also based on a time period theatingduring which the air dryer 104 has been in a heating phase. The heating phase time period theatingmay be determined based on heating phase control signals being provided by the air dryer control unit 270. As is mentioned above, the air dryer 104 is often provided with a heater 250. The drainage valve heater 250 is arranged at the drainage valve 230 to keep the valve warm, at least above freezing temperatures, such that water within the valve 230 does not freeze, whereby ice plugging of the drainage valve 230 is avoided. The time period theatingmay be compared to a time threshold theating_thresholdof a group of time thresholds and/or to a time interval theating_intervalof a group of time intervals. These time thresholds and/or time intervals may have values chosen depending on at which stage of the prediction algorithm the time thresholds and/or time intervals are to be utilized.

According to an embodiment of the present invention, the one or more parameters being related to vehicle use, and on which the prediction is based, include a parameter related to at least one time period tpressure_1,tpressure_2 ,... tpressure_nduring which the vehicle 100 has been in operation at an ambient pressure Pambientwithin at least one ambient pressure interval Pambient_interval_l, Pambient_interval_2,... Pambient_interval_nrespectively . Hereby, the prediction may at least partly take into account where, i.e. at which altitudes, and/or under which weather conditions the vehicle has been used, i.e. has been in operation. Different altitudes have different ambient pressures, as is well known. For example, for a vehicle being in operation at higher altitudes, there may be an increased risk for drainage valve problems compared to a vehicle being in operation at lower altitudes.

Essentially, any suitable number of time periods and ambient pressure intervals may be used as basis for the prediction. For example, four time periods related to four intervals may be used: - a first value tpressure_1indicating the time period the ambient pressure Pambienthas been within a first interval Pambient_interval_1delimited by 75 kPa and 80 kPa; 75 kPa < Pambient< 80 kPa; - a second value tpressure_2indicating the time period the ambient pressure Pambienthas been within a second interval Pambient_interval_2delimited by 85 kPa and 90 kPa; 85 kPa < Pambient< 90 kPa; - a third value tpressure_3indicating the time period the ambient pressure Pambienthas been within a third interval Pambient_interval_3delimited by 90 kPa and 95 kPa; 90 kPa < Pambient< 95 kPa; - a fourth value tpressure_4indicating the time period the ambient pressure Pambienthas been within a fourth interval Pambient_interval_4 being greater than 105 kPa, 105 kPa < Pambient· Also, other time periods related to other intervals for the ambient pressure Pambientmay of course be used as basis for the prediction. These intervals may be chosen based e.g. on characteristics of the drainage valve, such a manufacturer and/or other vehicle specifications, such as an engine size, a power take off, which axles/shafts being provided with air suspensions and/or a type of the vehicle, e.g. a truck, a timber vehicle, a van, a long distance bus or a city bus.

According to an embodiment of the present invention, the one or more parameters being related to vehicle use, and on which the prediction is based, include a parameter related to which type the vehicle 100 including the drainage valve 230 is.

These types of vehicles may include e.g. a truck and a bus. Also, other vehicle types may be indicated here, such as different types of trucks, e.g. a long distance haulage truck or a heavy load truck, or different types of cars. Different types of vehicles utilize pressurized air to different extents, wherefore the prediction may also be based on the type of the vehicle. For example, a bus type vehicle may utilize more pressurized air, e.g. for opening doors and/or for gear shifting, due to many starts and stops, than a long distance haulage truck, which may travel long distances at essentially one speed and one gear.

According to an embodiment of the present invention, the one or more parameters being related to vehicle use, and on which the prediction is based, include a parameter related to a time period tvelocityduring which the vehicle 100 including the drainage valve 230 has had a speed v within a predetermined velocity interval vinterval ·This predetermined velocity interval vintervalmay be chosen such that it can be used for identifying a specific use of the vehicle. One such non-limiting example interval is delimited by 77 km/h and 83 km/h; i.e. 77 km/h < vinterval< 83 km/h; which may be related to typical speed for a truck vehicle travelling on a highway in Europe. Also, other predetermined intervals may be used, depending e.g. on a type of the vehicle and/or on speed regulations in the geographical area where the vehicle is in operation.

As mentioned above, the one or more parameters on which the drainage malfunction prediction is made may include a suitable number of parameters. Different parameters make different contributions to the prediction, e.g. in terms of reliability and exactness for the prediction. According to an embodiment of the present invention, the parameters may be classified regarding their importance for the prediction quality. Hereby, a suitable number of parameters to be included in the prediction may be selected based on the classification.

As an example, the following importance classification for some above mentioned parameters may be used, where number 1 is the parameter being classified as contributing most to the reliability of the prediction, and parameters of increasing numbers contribute less and less: 1.A time period ttempthe vehicle has been in operation at an ambient temperature Tambientwithin a predefined temperature interval Tambient_interval; 2.A distance D travelled by the vehicle; 3.A time period tregenerationthe air dryer has been in a regeneration phase; 4.A time period theatingthe air dryer has been in a heating phase ; .A first value tpressure_1indicating the time period the ambient pressure Pambienthas been within a first interval Pambient_interval_1delimited by 75 kPa and 80 kPa; 75 kPa < Pambient< 80 kPa; 6.A second value tpressure_2indicating the time period the ambient pressure Pambienthas been within a second interval Pambient_interval_2delimited by 85 kPa and 90 kPa; 85 kPa < Pambient< 90 kPa; 7.A third value tpressure_3indicating the time period the ambient pressure Pambienthas been within a third interval Pambient_interval_3delimited by 90 kPa and 95 kPa; 90 kPa < Pambient< 95 kPa; 8.A fourth value tpressure_4indicating the time period the ambient pressure Pambienthas been within a fourth interval Pambient_interval_4being greater than 105 kPa; 105 kPa < Pambient · 9 The type of the vehicle; . A time period tvelocitythe vehicle has had a speed v within a predetermined velocity interval vinterval; e.g. 77 km/h < Vinterval< 83 km/h.

Generally, the more parameters the prediction is based on, the more reliable the prediction is. Therefore, a wanted reliability for the prediction may be chosen, e.g. by starting with the most important parameter according to the classification, e.g. parameter 1 above related to the time period ttemp, and then add parameters from the classification, possibly by starting with adding the second most important parameter, e.g. parameter 2 above related to the distance D, and so on, until a wanted/needed reliability for the prediction is reached.

Generally, the values, thresholds and/or intervals mentioned for time periods, temperatures, pressures and/or distances in this document may include absolute values, percentages and/or ratios .

The herein used thresholds and/or intervals may be set and/or predefined based on operational vehicle data, such as historic operational vehicle data. Basically, operational vehicle data may be collected and stored during operation of the vehicle. Then, this data may be analyzed and correlated to exchange, malfunction and/or service for the drainage valve 230. Hereby, the values for the thresholds and/or the intervals may be adapted based on the historic operational data such that drainage valves being probable to malfunction within the prediction time period tpredmay be reliably identified.

If the air dryer 104 and/or the drainage valve 230 are exchanged and/or serviced, one or more of the parameters, e.g. the time period parameters, are according to an embodiment reset to zero, such that the time periods are calculated for a use of a new drainage valve individual and/or for a use of a serviced drainage valve individual, respectively.

According to an embodiment of the present invention, the prediction of the probability for a drainage valve malfunction is determined by usage of one or more algorithms, possibly also being useful within machine learning systems, resulting in a decision/conclusion indicating either of: - "Yes, a drainage valve malfunction within the upcoming prediction time period tpredis probable"; and - "No, a drainage valve malfunction within the upcoming prediction time period tpredis not probable".

This decision/conclusion may according to an embodiment be calculated using a decision tree algorithm. The decision tree algorithm includes a flowchart-like structure having a number of internal nodes, each node representing a test, such as e.g. a comparison of a measured value with a threshold and/or an interval. From the nodes, two branches extend, each branch representing one outcome of the test. By these internal nodes and branches, a tree structure is built, and the leaves at the end of each last branch indicates a decision, i.e. if it is probable ("Yes") or if it is not probable ("No") that a drainage valve malfunction will occur within the upcoming prediction time period tpred· The decision/conclusion may according to an embodiment also be determined/calculated by use of a generalized boosted regression model (GBM) algorithm. The GBM algorithm is a machine learning technique for regression and classification problems, including e.g. an ensemble of decision trees, which is optimized based on a cost and/or loss function. The GBM algorithm provides a decision, i.e. if it is probable ("Yes") or if it is not probable ("No") that a drainage valve malfunction will occur within the upcoming prediction time period tpred· According to an embodiment of the present invention, a problem related to missing data, e.g. due to a broken sensor, may be solved by entering a default/predetermined value into the decision making algorithms. Thus, the prediction of the drainage valve malfunction probability according to the herein described embodiments may be performed also if there is some data missing. The default/predetermined values may e.g. be chosen as a mean value for that parameter for the vehicle, and/or for a number of vehicles, such as e.g. a fleet of vehicles .

A person skilled in the art will appreciate that a method for prediction of a probability for a malfunction of a drainage valve according to the present invention can also be implemented in a computer program, which, when it is executed in a computer, instructs the computer to execute the method. The computer program is usually constituted by a computer program product 403 (figure 4) stored on a non-transitory/nonvolatile digital storage medium, in which the computer program is incorporated in the computer-readable medium of the computer program product. Said computer-readable medium comprises a suitable memory, such as, for example: ROM (Read-Only Memory), PROM (Programmable Read-Only Memory), EPROM (Erasable PROM), Flash memory, EEPROM (Electrically Erasable PROM), a hard disk unit, etc.

Figure 4 shows in schematic representation a control unit 400. The control unit 400 comprises a computing unit 401, which can be constituted by essentially any suitable type of processor or microcomputer, for example a circuit for digital signal processing (Digital Signal Processor, DSP), or a circuit having a predetermined specific function (Application Specific Integrated Circuit, ASIC). The computing unit 401 is connected to a memory unit 402 arranged in the control unit 400, which memory unit provides the computing unit 401 with, for example, the stored program code and/or the stored data which the computing unit 401 requires to be able to perform computations. The computing unit 401 is also arranged to store partial or final results of computations in the memory unit 402.

In addition, the control unit 400 is provided with devices 411, 412, 413, 414 for receiving and transmitting input and output signals. These input and output signals may comprise waveforms, impulses, or other attributes which, by the devices 411, 413 for the reception of input signals, can be detected as information and can be converted into signals which can be processed by the computing unit 401. These signals are then made available to the computing unit 401. The devices 412, 414 for the transmission of output signals are arranged to convert signals received from the computing unit 401 in order to create output signals by, for example, modulating the signals, which can be transmitted to other parts of and/or systems in the vehicle.

Each of the connections to the devices arranged for receiving and transmitting input and output signals may be include one or more of a cable; a data bus, such as a CAN bus (Controller Area Network bus), a MOST bus (Media Orientated Systems Transport bus), or some other bus configuration; or a wireless connection. A person skilled in the art will appreciate that the above-stated computer may comprise the computing unit 401 and that the above- stated memory may include the memory unit 402.

Control systems in modern vehicles commonly comprise communication bus systems including one or more communication buses for linking a number of electronic control units (ECU's), or controllers, and various components located on the vehicle. Such a control system may comprise a large number of control units and the responsibility for a specific function may be divided amongst more than one control unit. Vehicles of the shown type thus often comprise significantly more control units than are shown in figure 1, which is well known to the person skilled in the art within this technical field .

In the shown embodiment, the present invention is implemented in the control unit 400. The invention can also, however, be implemented wholly or partially in one or more other control units 160, 170, 180, 190, 270 already present within the vehicle and/or external to the vehicle, or in some control unit 160, 170, 180, 190, 270 dedicated to the present invention .

Here and in this document, units are often described as being arranged for performing steps of the method according to the invention. This also includes that the units are designed to and/or configured to perform these method steps.

These method steps, and their corresponding units, may for example correspond to groups of instructions, which can be in the form of programming code, that are input into, and are utilized by a processor when the steps/units are active and/or are utilized for performing the method step, respectively.

According to an aspect of the present invention, a system arranged for prediction of a probability for a malfunction of a drainage valve 230 of an air dryer 104 is presented. The air dryer 104 is, as mentioned above, included in a pressurized air system 150 of a vehicle 100, and is used for discharging humidity being absorbed from pressurized air flowing through the air dryer 104.

The system includes a control unit 170, 180, which is arranged for determining 302 the drainage valve malfunction probability prediction such that it indicates the probability for a malfunction during an upcoming prediction time period tpredof predefined length, e.g. 15 days.

The control unit 170, 180 is further arranged for determining 302 the drainage valve malfunction probability prediction based at least on one or more parameters related to a use of the vehicle 100, as is described in detail above.

The system may be configured in a large number of ways, and the calculation/determination of the prediction may be performed in a number of different locations, internally and/or externally of the vehicle.

According to an embodiment of the present invention, the control unit 170 arranged for determining the drainage valve malfunction probability is arranged as included within the vehicle 100, as illustrated in figure 1. The control unit 170 is then provided with the one or more parameters being needed for determining the prediction from a second control unit of the vehicle, such as e.g. an air system control unit 160, an air dryer control unit 270, a central control unit of the vehicle 190, a distributed control unit of the vehicle, and/or from one or more internal and/or external sensors 191 of the vehicle 100, such as e.g. an ambient temperature sensor, an ambient pressure sensor and/or a vehicle velocity/speed sensor. The control unit 170 then determines/calculates the drainage valve malfunction probability based on the received parameter data, e.g. by usage of a suitable evaluation software implementing the method steps herein described.

When the internal control unit 170 determines the predictions of the drainage valve malfunction probability, the information related to these predictions may be sent from the vehicle to a node external from the vehicle, e.g. to a fleet management node 181, whereby a wireless connection 183, for example according to a suitable wireless communication standard may be utilized for the transfer of the data.

The fleet management node 181 performs an analysis of the prediction information and may then suggest a suitable action to a vehicle owner 182. This action may be to suggest taking the vehicle to a service station to check the status of the air dryer 104 and/or drainage valve 230.

According to another embodiment of the present invention, the control unit 180 arranged for determining the drainage valve malfunction probability is arranged outside/external of the vehicle 100 and is provided with the one or more parameters from a second communication control unit 170 on the vehicle 100. The external control unit 180 is provided with the one or more parameters by the communication control unit 170 via some kind of data carrying link 183, which may include one or more cables and/or wireless communication equipment. For example, the one or more parameters may be transmitted from the communication control unit 170 during service, whereby a cable and/or wireless connection 183 may be used for the transfer of the data. Also, the one or more parameters may be transmitted from the communication control unit 170 to the control unit 180 during normal operation of the vehicle, whereby a wireless connection 183, for example according to a suitable wireless communication standard may be utilized for the transfer of the data .

When the control unit 180 is located externally from the vehicle, the control unit 180 determines/calculates the drainage valve malfunction probability prediction based on the received parameter data. Hereby, a suitable evaluation software implementing the method steps herein described may be used for calculating the predictions. When the external control unit 180 has determined the predictions of the drainage valve malfunction probability, the information related to these predictions may be sent to a fleet management node 181, which performs an analysis of the prediction information. The fleet management node may then suggest a suitable action to a vehicle owner 182.

The system according to the present invention may be arranged for performing all of the above, in the claims, and in the herein described embodiments method steps. The system is hereby provided with the above described advantages for each respective embodiment.

A skilled person also realizes that the above described system may be modified according to the different embodiments of the method of the present invention. The present invention is also related to a vehicle 100, such as a truck, a bus or a car, including the herein described system for prediction of a probability for a malfunction of a drainage valve.

The present invention is not limited to the above described embodiments. Instead, the present invention relates to, and encompasses all different embodiments being included within the scope of the independent claims.

Claims

1. Method for prediction of a probability for a malfunction of a drainage valve (230) of an air dryer (104) included in a pressurized air system (150) in a vehicle (100), said drainage valve (230) being arranged for discharging humidity being absorbed from pressurized air (230) flowing through said air dryer (104); characterized in that - said predictionindicates said probability for said malfunction during an upcoming prediction time period tpredof predefined length; and - said prediction being based at least on one or more parameters being related to a use of said vehicle (100), wherein said one or more parameters include a parameter related to a time period ttempduring which said vehicle (100) including said drainage valve (230) has been in operation at an ambient temperature Tambientwithin a predefined temperature interval Tambient_interval, and/or wherein said one or more parameters include a parameter related to at least one time period tpressure_1,tpressure_2,... tpressure_nduring which said vehicle (100) has been in operation within at least one ambient pressure interval Pambient_interval_l, Pambient_interval_2, ... Pambient_interval_nrespectively.

2. The method as claimed in claims 1, wherein said one or more parameters include a parameter related to a distance D travelled by said vehicle (100) including said drainage valve (230) .

3. The method as claimed in any one of claims 1-2, wherein said one or more parameters include a parameter related to a type of said vehicle (100) including said drainage valve (230) .

4. The method as claimed in any one of claims 1-3, wherein said one or more parameters include a parameter related to a time period tvelocityduring which said vehicle (100) including said drainage valve (230) has travelled in a predetermined velocity interval vinterval.

5. The method as claimed in any one of claims 1-4, wherein said prediction is also based on a time period tregenerationduring which said air dryer (104) has been in a regeneration phase.

6. The method as claimed in any one of claims 1-5, wherein said prediction is also based on a time period theatingduring which said air dryer (104) has been in a heating phase.

7. The method as claimed in any one of claims 1-6, wherein said prediction is determined using one or more of: - a decision tree algorithm; and - a Generalized boosted regression model (GBM) algorithm.

8. The method as claimed in any one of claims 1-7, wherein said upcoming prediction time period tpredis 15 days.

9. Computer program, characterized in code means, which when run in a computer causes the computer to execute the method according to any of the claims 1-8.

10. Computer program product including a computer readable medium and a computer program according to claim 9, wherein said computer program is included in the computer readable medium.

11. A system arranged for prediction of a probability for a malfunction of a drainage valve (230) of an air dryer (104) included in a pressurized air system (150) of a vehicle (100), said drainage valve (230) being arranged for discharging humidity being absorbed from pressurized air flowing through said air dryer (104); characterized by - a control unit (170, 180) arranged for - determining (302) said prediction such that it indicates said probability for said malfunction during an upcoming prediction time period tpredof predefined length; and - determining (302) said prediction based at least on one or more parameters being related to a use of said vehicle (100), wherein said one or more parameters include a parameter related to a time period ttempduring which said vehicle (100) including said drainage valve (230) has been in operation at an ambient temperature Tambientwithin a predefined temperature interval Tambient_interval, and/or wherein said one or more parameters include a parameter related to at least one time period tpressure_1, tpressure_2,...tpressure_nduring which said vehicle (100) has been in operation within at least one ambient pressure interval Pambient_interval_1, Pambient_interval_2, ... Pambient_interval_nrespectively.

12. System as claimed in claim 11, wherein said control unit (180) is located outside said vehicle (100) and is provided with at least said one or more parameters from a second control unit (170) on said vehicle (100).

13. System as claimed in claim 11, wherein said control unit (170) is located within said vehicle (100) and is provided with at least said one or more parameters from a second control unit of said vehicle and/or from one or more sensors of said vehicle (100).