The present invention generally relates to the field of detection of leakages in fluid supply networks by means of detecting abnormal pressure.

Pipeline networks are the most economic and safest mode of transportation and distribution for fluids like water, oil, gases and other fluid products. As a means of long-distance transport, pipelines have to fulfill high demands of safety, reliability and efficiency. If properly maintained, pipelines can last indefinitely without leaks. The leaks in the pipeline network can be caused by various reasons: damage from nearby excavation equipment, corrosion of pipes, accidents, earth movement etc. Such a system of interconnected pipes that carries a pressurized fluid such as water, oil, gases and other fluid products is called hydraulic supply network. When monitoring hydraulic supply networks, one often faces the task of abnormal state detection. Since such abnormal state of the fluid supply network, for example abnormal pressure in the fluid supple network, can be an evidence of a problem in the fluid supply network, e.g. of a leakage. If there is a problem in the network, it is very important to troubleshoot the defect in short time. Timely localization of the problem allows reducing cost of repairing and possible liquid losses on the network. Consequences of the leakage can be very destructive .

Any fluid supply network, including a water supply system, typically includes: fluid sources or fluid storage facilities such as reservoirs, tanks, etc.; pressurizing components such as pumping stations, pumps, etc . ; a network of pipes for transportation and distribution of fluid to consumers.

Further the fluid supply network is considered on an example of a water supply network. A water supply system or water supply network also belongs to the hydraulic supply networks which provide water supply to different types of consumers.

The water in the supply network is maintained at positive pressure to ensure that water reaches all parts of the network, that a sufficient flow is available at every take-off point, i.e. at every consumer, and to ensure that untreated water in the ground cannot enter the water supply network. The water is typically pressurized by the pressurizing components. Different types of pipes are used in the pipe network of the water supply network. In general, the pipes can be classified in two categories depending on purpose : main pipes or transportation pipes, that are mainly long pipes located underground with large diameters of, for example, 300 - 700 mm, but can be of giant diameters of more than 3m, moving pressurized water from the water storage facilities into the town or district of the town. distribution pipes, that are pipes with small diameters of, for example, 80 - 300 mm, used to take the water from the main pipes to the consumers, which may be private houses as whole or each apartment individually, or industrial, commercial or institution establishments, and other usage points such as fire hydrants.

By now water has become one of the most important goods in the 21st century. However, sometimes considerable water losses occur in water supply networks. In many cases, minor water leakages deriving from the inefficient hydraulic seal of junctions or from small cracks on pipes may lie hidden for a long time, sometimes for months or even years. Major leakages can be easily observed when significant damages to the pipes occur, as they usually result in large amounts of water erupting from ground or flowing in the consumer properties.

The proven method around the world to reduce leakage from the water supply system is to proactively find the leaks before they appear at the surface. This can be achieved by monitoring the state of the fluid supply network and has the benefits of reducing the time the leaks are running, and wasting water.

The problem of the leakage detection can be solved using the local leak detection methods by using various equipment, such as acoustic equipment, thermograph, ground penetration radar, etc. In most cases such method requires a lot of labor since experts with the respective equipment should trace the whole supply network to find the leakage. Also there are global methods, including methods on detecting abnormal state in the network, that are based on gathering a lot of data about the network condition such as fluid pressure and pipe flows and analyzing them. Sometimes gathering such data is connected with the installation of the expensive equipment like flow meters, especially in case such flow meters have to be installed on to the main pipes.

One of the most important stages of global methods is the fluid supply network modeling. Model of the fluid supply network must be representative, i.e. represent the real network as accurately as possible. However in most cases it is not a trivial task. Very often there is no information about some of the network parameters, e.g. the pipeline roughness or vertical topology of the fluid supply network can be unknown.

In general, the existing ways of leakage detection in a fluid supply network require very complicated calculations and / or very sophisticated and expensive equipment the installation of which is complex, costly and time consuming.

In the light of the foregoing discussion, it is evident that there is a strong need of easy and convenient abnormal state detection in the fluid supply network.

The object is solved by a method for abnormal state detection in a fluid supply network as defined in claim 1, and by a system for abnormal state detection in a fluid supply network as defined in claim 8. Consequently, the present invention provides a method for detecting abnormal state in a fluid supply network based on pressure patterns analysis, where the fluid supply network comprises a network of pipes for delivering a fluid to consumers and at least one pressure sensor. The at least one sensor is installed in a pipe of the networks of pipes and adapted to make a measurement of fluid pressure in that pipe where the pressure sensor is installed. According to the present invention the method comprises at least a training mode and at least a test mode. Within the training mode, the historical pressure data with the at least one pressure sensor are collected by taking a plurality of measurements of the fluid pressure measured by the at least one pressure sensor. And further on at least one historical fluid pressure behavior portrait for the at least one pressure sensor is created based on the collected historical pressure data.

Within the test mode actual pressure data with the at least one sensor are measured by taking at least one measurement of the fluid pressure measured by the at least one pressure sensor, and further compared with the at least one historical fluid pressure behavior portrait which was created within the training mode. After that the judgment about the presence of abnormal state of the fluid supply network is made. An abnormal state is judged to be present when the measured actual pressure data is not within the historical fluid pressure behaviour portrait.

Furthermore, the present invention provides a system for detecting abnormal state of a fluid supply network based on pressure patterns analysis. According to the present invention the system comprises a plurality of pressure sensors wherein each pressure sensor is adapted to be located on a pipe of the network of pipes and adapted to make a measurement of fluid pressure in the pipe where the pressure sensor is installed.

Moreover the system comprises a monitoring unit configured to perform pressure pattern analysis in accordance with a method of any one of the claims 1 - 5.

Therefore there is a fluid supply network with the system for detecting abnormal state of a fluid supply network based on pressure pattern analysis.

The present invention is based on the insight that emergence of the fluid leakage somewhere in the fluid supply network will bring to the pressure drop in the pipe where the leakage is appeared and therefore an abnormal state of the fluid supply network can be detected by the pressure sensors installed on the pipes. In addition to that it is possible to create a historical portrait of fluid pressure behavior for each pressure sensor by collecting historical pressure data with the respective pressure sensor.

Such historical fluid pressure behavior portrait can be interpreted as a normal behavior of the fluid supply network in the pipe where the respective pressure sensor is installed. Such assumption is based on the insight that the historical fluid pressure behaviour portrait is created based on collected historical pressure data that are aggregated in the way when abnormalities are eliminated by such aggregation. Therefore in case actual pressure data measured by the respective pressure sensor is not within the historical fluid pressure behavior portrait created for this particular pressure sensor the abnormal pressure, and consequently, the abnormal state of the fluid supply network is detected and the fluid supply network is in abnormal state in the particular location where the respective pressure sensor is installed. Therefore the judgment of a presence of abnormal state of the fluid supply network can be made based on pattern analysis. Moreover a leakage in the particular location of the respective pressure sensor can be suspected.

This method for detecting abnormal state of the fluid supply network does not require installation of expensive equipment, for example such as flow meters as it was described above Thus, the present invention is proposed to provide a new method and a system for detecting abnormal state in the fluid supply network.

Further embodiments of the present invention are subject of the further sub-claims and of the following description, referring to the drawings.

In a possible embodiment the at least one historical fluid pressure behavior portrait which is created for the at least one pressure sensor based on the collected historical data within the training mode comprises an average of the collected historical pressure data for a particular point in time and a statistical deviation of the collected historical pressure data that characterizes the variability of data in the corresponding point in time.

In each particular point of time the value of pressure taken on the corresponding pressure sensor can differ from day to day within the given time period the historical pressure data are collected. Therefore the historical pressure data for the each particular point of time collected within the given time period should be aggregated to create the historical fluid pressure behaviour portrait for the respective pressure sensor. An aggregation function could be any function, which transforms vector into scalar value, for example a simple arithmetic average function. In addition to that the statistical deviation of the collected historical pressure data should be represented on the historical fluid pressure behavior portrait.

In preferable case the aggregation function for calculating the average of the collected historical pressure data and for calculating the statistical deviation of the collected historical pressure data should correspond to each other. For example, if a simple average function is used to calculate the average of the collected historical pressure data, the average absolute deviation should be used to calculate the statistical deviation of the collected historical pressure data.

The above said feature allows creating more precise historical fluid pressure behaviour portrait due to eliminating noises and abnormalities. In an enhanced embodiment a median function is used to calculate the average of the collected historical pressure data for the particular point in time to create the at least one historical fluid pressure behavior portrait for the at least one pressure sensor based on the collected historical data. And the statistical deviation of the collected historical pressure data is calculated as a median absolute deviation function.

The median function and median absolute deviation function allow smoothing the accidental spikes and other shortcomings in the collected historical pressure data.

In a possible embodiment the step of judging the presence of abnormal state of the fluid supply network includes estimating an intensity of pressure abnormality based on the calculated pressure data difference between the historical fluid pressure behavior portrait and the measured actual pressure data.

This feature allows to avoid false-positive results where the

"positive" statement if the abnormal condition presence. In a possible embodiment the at least one pressure sensor is installed on the distribution pipe.

As it is mentioned above a network of pipes for delivering a fluid to consumers comprises at least one main pipe to transport fluid from a source into the fluid supply network, and at least one distribution pipe, wherein the at least one distribution pipe is adapted to transport fluid from the main pipe to a consumer fluidly connected to the distribution pipe . The distribution pipes are pipes with small diameter (e.g. less than 300 mm) to carry the pressurized fluid from the main pipe to the consumers . The pressure sensors that are installed on the distribution pipes are relatively cheap in comparison with the ones that installed on the main pipes. In addition to that the installation process of pressure sensors on a distribution pipe is not connected with the interruption of fluid supply to the consumers of the fluid supply network.

Therefore using pressure sensors installed on the distribution pipes makes this method more cost effective in comparison with the case when pressure sensors are installed on the main pipe .

In other possible embodiment the pressure of the fluid provided by a supply network pressurizing component to the network of pipes is also measured and taken into account. As it is mentioned above the fluid supply network further comprises at least one supply network pressurizing component that is adapted to provide a predefined positive pressure in the network of pipes, wherein the fluid supply network is further equipped with a pumping station pressure sensor.

The pumping station pressure sensor is installed downstream of the at least one supply network pressurizing component and there is no pipe junctions between the at least one supply network pressurizing component and the pumping pressure sensor. The pumping station pressure sensor is adapted to make a measurement of pressure of the fluid provided by the supply network pressurizing component to the network of pipes. Wherein the at least one supply network pressurizing component is providing a predefined positive pressure in the network of pipes .

Therefore within the training mode at the step of collecting historical pressure data, the historical pressure data are further collected with the pumping station pressure sensor by taking a plurality of measurements of the fluid pressure provided by the at least one supply network pressurizing component .

In addition to that between the step of collecting historical pressure data and the step of creating the at least one historical fluid pressure behavior portrait, the method comprises a step on which the collected historical pressure data are calculated as a difference between the pressure measured by the at least one pressure sensor and the pressure measured by the pumping station pressure sensor for each point in time. Therefore the at least one fluid pressure behavior portrait is created based on the collected historical pressure data which are calculated just said above . Within the test mode at the step of measuring the actual pressure data, the actual pressure data are further measured on the pumping station pressure sensor.

In addition to that between the step of measuring the actual pressure data and the step of comparing the measured actual pressure data with the at least one historical fluid pressure behavior portrait, the method comprises a step of calculating the measured actual pressure data which is calculated as a difference between the actual pressure data measured by the at least one pressure sensor and the actual pressure data measured by the pumping station pressure sensor.

This embodiment is sensitive to the changes of pressure maintained by the supply network pressurizing component.

In other possible embodiment the fluid supply network is a water supply network.

In possible embodiment the system further comprises at least one pumping station pressure sensor that is installed downstream of the at least one supply network pressurizing component and there is no pipe junctions between the at least one supply network pressurizing component and the pumping pressure sensor. Such pumping station pressure sensor is adapted to make a measurement of pressure of the fluid provided by the supply network pressurizing component to the network of pipes. Moreover the system comprises the monitoring unit configured to perform pressure pattern analysis in accordance with method of the claim 6.

For a more complete understanding of the present invention and advantages thereof, reference is now made to the following description taken in accompanying drawings. The invention is explained in more detail below using exemplary embodiments which are specified in the schematic figures of the drawings, in which:

Fig. 1 shows a block diagram of a fluid supply network;

Fig. 2 shows a flow diagram of a method for detecting abnormal state of a fluid supply network according to the present invention;

Fig. 3 shows an example of the fluid pressure measurements taken by one pressure sensor; Fig. 4 shows an example of a historical fluid pressure behaviour portrait for the respective pressure sensor;

Fig. 5 shows an example of historical fluid pressure behaviour portraits of the business days and of the weekends; Fig. 6 shows an example of normalized pressure data difference;

Fig. 7 shows a block diagram of another embodiment of the method for detecting abnormal state of a fluid supply network according to the present invention; Fig. 8 shows a block diagram of a system for detecting abnormal state of a fluid supply network according to the present invention;

Fig. 9 shows a block diagram of another embodiment of the system for detecting abnormal state of a fluid supply network according to the present invention;

Various embodiments are described with reference to the drawings, wherein like reference numerals are used to refer to like elements throughout. In the following description, for purpose of explanation, numerous specific details are set forth in order to provide a thorough understanding of one or more embodiments. It may be evident that such embodiments may be practice without these specific details .

The invention relates to a system 28 for detecting abnormal pressure in a fluid supply network 1 based on pressure patterns analysis.

FIG 1 illustrates a water supply network 1 with a water source (not shown in FIG 1) , a supply network pressurizing component 2, main pipes 3 and distribution pipes 4. The fluid pressurized by the supply network pressurizing component 2 is being transported through main pipes 3. Distribution pipes 4 are used to connect the consumers 5 with the main pipes 3. The main pipes 3 and distribution pipes 4 might be equipped with pressure sensors 6. Such pressure sensors 6 can be installed directly in the main pipes 3, or in the distribution pipes 4. As it is mentioned above said distribution pipes 4 are pipes with small diameter (e.g. less than 300 mm) to carry the pressurized fluid from the main pipe 3 to the consumers 5. The pressure sensors 6 that are installed on the distribution pipes 4 are relatively cheap in comparison with the ones that are installed on the main pipes 3. In addition to that the installation process of said pressure sensors 6 on the distribution pipes is not connected with the interruption of fluid supply to the consumers 5.

The more pressure sensors 6 are installed in the fluid supply network 1 the more complete picture of abnormal pressure in the fluid supply network 1 can be done, therefore further leakage detection & localization can be accomplished.

The water supply network also can be equipped with a pumping station pressure sensor 7, that is installed downstream of the supply network pressurizing component 2 and there is no pipe junctions between the at least one supply network pressurizing component and the pumping station pressure sensor 7. The pumping station pressure sensor 7 is adapted to make a measurement of pressure of the fluid provided by the supply network pressurizing component 2 to the network of pipes .

FIG. 2 shows a flow diagram of an embodiment of a method for detecting abnormal pressure in a fluid supply network 1 based on pressure pattern analysis according to the present invention. The method comprises at least a training mode I and at least a test mode II. The training mode can be processed also in off-line regime.

Within the training mode I at step 8 the historical pressure data that comprise plurality of the fluid pressure measurements taken by the at least one pressure sensor 6 are collected.

The historical pressure data are collected within a given time period with a given time step. The given time period and the given time step can be defined by experts. The given time period to collect the historical pressure data can be, for example, 1 month or 2 months or 1 year. The given time step can be, for example, 5 or 15 minutes. However the less given time step is the more precise the historical fluid pressure behavior portrait can be created on further steps. On the other hand the size of the given time step can be limited by an equipment, e.g. by a pressure sensor.

The example of the fluid pressure measurements taken by the pressure sensor 6 is shown on FIG 3. The axis 9 on the FIG3 is time, while the axis 10 is pressure. There may be present various kind of measurement noise such as gaps 11, zero- values, spikes, etc, generally characterized by an abnormal value in the sampled continuous data stream. Therefore to obtain better results pre-processing of the fluid pressure measurements such as for example gap- filling and / or boundaries cut-offs can be performed to eliminate the measurement noise.

At step 12 at least one historical fluid pressure behaviour portrait for the at least one pressure sensor 6 is created. The example of the historical fluid pressure behaviour portrait 13 for the respective pressure sensor 6 is shown on FIG 4. The axis 14 is time. The data interval for the axis 14 is one day. The axis 15 is the value of pressure.

The historical fluid pressure behavior portrait 13 for the particular pressure sensor 6 is a day graph 16 created based on aggregation of the collected historical pressure data received with this particular pressure sensor 6 within the given time period with the given time step. In other words for each point in time within a day all historical pressure data measured by the respective pressure sensor 6 in this particular point in time during the given time period should be aggregated. In enhanced embodiment of the method the aggregation of the collected historical pressure data can be done with the help of any average function which transforms vector into scalar value, for example, an arithmetic average function.

In addition to that a statistical deviation 17 of the collected historical pressure data for each point in time also should be estimated and added to the historical fluid pressure behaviour portrait 13. The statistical deviation 17 characterizes the variability of collected historical pressure data in the corresponding point in time within the given time period of collecting historical pressure data. Moreover in the ideal case the function applied to calculate the statistical deviation 17 should correlate with the function that is used to aggregate the collected historical pressure data to create the graph 16. In case of arithmetic average function is applied to the collected historical pressure data to create the graph 16, the arithmetic absolute deviation should be calculated to define the statistical deviation 17.

In the enhanced embodiment of the method the median function and median absolute deviation are used to aggregate the collected historical pressure data to create the at least one historical fluid pressure behaviour portrait 13. Such aggregation allows smoothing the accidental spikes and other shortcomings in the collected historical pressure data. Therefore it can be assumed that such historical fluid pressure behaviour portrait 13 is a portrait of the fluid pressure in the fluid supply network 1 wherein there are no leakages in the fluid supply network 1.

However because of possible different behaviour of the consumers and therefore different consumption within the given time period a plurality of historical fluid pressure behaviour portraits 13 can be created for each pressure sensor 6 to reflect different fluid pressure behaviour within different periods, for example within different periods within a week. For example the historical fluid pressure behaviour portraits of the business days and of the weekends can be created. FIG 5 presents two historical fluid pressure behaviour portraits - one is for the business days (Monday to Friday) 18 and another one is for weekend days 19. However other periods with specific fluid consumption and respectively specific fluid pressure behavior, for example based on seasons, and / or holidays, can be defined by experts to create historical fluid pressure behaviour portraits . In preferable case at least one historical fluid pressure behaviour portrait 13 should be created for each sensor 6. However it is reasonable to create plurality of historical fluid pressure behaviour portraits 13 for each sensor 6, at least two of them. The number of the historical fluid pressure behavior portraits should be defined by experts and by analyzing the collected historical pressure data.

Therefore after performing the training mode I there is at least one historical fluid pressure behavior portrait 13 created for the at least one pressure sensor 6 that is installed on the pipes 3, 4 of the fluid supply network 1.

Within the test mode II at step 20 the measurement of the actual fluid pressure on the at least one pressure sensor 6 is taken.

At step 21 the measured actual pressure data is compared with the respective historical fluid pressure behavior portrait 13 created for the respective pressure sensor 6 at step 12.

At step 22 the presence of abnormal state of the fluid supply network 1 based on the result of the step 21 is judged. Wherein abnormal state of the fluid supply network is judged to be present when the measured actual pressure data is not within the historical fluid behavior pressure portrait 13.

In the enhanced embodiment of the method the step of judging 22 the presence of abnormal state of the fluid supply network 1 includes calculating a pressure data difference between the historical fluid pressure behavior portrait 13 and the measured actual pressure data and further estimating an intensity of pressure abnormality based on the calculated pressure data difference. In addition to that before estimating a pressure data difference the calculated pressure data difference can be normalized by the statistical deviation 17. Such normalization can be done, for example, by using Z-score approach with the following general equation: z =—σ- , where

x is the measured actual pressure data measured;

μ is the respective pressure data from the historical fluid pressure behavior portrait 13 in the respective time point when the actual pressure data is measured on the respective pressure sensor 6;

σ is the width of the historical fluid pressure behavior portrait, i.e. the statistical deviation 17 of the data on the historical fluid pressure behavior portrait in the respective time point when the actual pressure data is measured on the respective pressure sensor 6.

FIG 6 illustrates an example of the normalized pressure data difference calculated at the step 22 as described above. The axis 22 is the time, while the axis 23 is the normalized pressure data difference. The peak 24 on Fig. 6 means that the measured actual pressure on the respective pressure sensor 6 at the moment of time 25 was 30 times more than normal, usual pressure at this period of time that is shown at the historical fluid pressure behavior portrait 13. Therefore with certainty the actual pressure measured at the moment of time 25 is abnormal, and consequently the state of the fluid supply network at the moment of time 25 is abnormal .

Such approach can support the process to estimate the intensity of a pressure abnormality that can be done by experts. In addition to that such approach allows avoiding false results.

In the enhanced embodiment of the method the at least one pressure sensor 6 is installed on the distribution pipe 4. Indeed the method according to the present invention allows detecting abnormal state of the fluid supply network 1 when all pressure sensors 6 are installed on the distribution pipes 4. Therefore there is no need in installing pressure sensors 6 on the main pipes 3.

FIG 7 illustrates the enhanced embodiment of the method that is applied when the at least one supply network pumping station 2 does not hold fluid pressure value on a specified level within the whole period of time, i.e. the fluid pressure provided by the supply network pumping station 2 varies from time to time within the training mode and within the test mode.

In such situation the abnormal behavior of the pressure on the pressure sensor 6 can be caused by the dropdown of the pressure provided by the supply network pumping station 2 , not by the fluid leakage. Such situation will lead to a false positive alarm.

Therefore to avoid such false positive alarm within the training mode I at step 8 in addition to the historical pressure data that comprises plurality of the fluid pressure measurements by the at least one pressure sensor 6 the historical pumping station pressure data are collected by taking plurality of the fluid pressure measurements by a pumping station pressure sensor 7. While the at least one supply network pumping station 2 provides a predefined positive pressure in the network of pipes. After that at step 26 the collected historical data are calculated as a difference between the pressure measured by the at least one pressure sensor 6 and the pressure measured by the pumping station pressure sensor 7 for each point in time .

Further at the step 12 the at least one historical fluid pressure behavior portrait 13 is created based on the collected historical data that are calculated within the step 26. It allows taking into account the possible fluctuation of fluid pressure provided by the at least one supply network pressurising component 2 at the creation of the at least one historical fluid pressure behavior portrait 13.

Within the test mode II at step 20 in addition to the measurement of actual pressure data of the fluid pressure on the at least one pressure sensor 6, the actual pressure data are further measured on the pumping station pressure sensor 7.

After that at step 27 the measured actual pressure data are calculated as a difference between the actual pressure data measured by the at least one pressure sensor 6 and the actual pressure data measured by the pumping station pressure sensor 7.

Further at step 21 the measured actual pressure data calculated within the step 27 is compared with the respective historical fluid pressure behavior portrait 13.

Therefore at the step 22 the judgement of the presence of abnormal state of the fluid supply network 1 is performed based on a result of the step 21 wherein the possible fluctuation of pressure provided by the at least one supply network pressurising component 2 are taken into account.

The method in accordance with the present invention can be applied as well to the fluid supply network 1 such as a water supply network. FIG 8 shows a system 28 for detecting abnormal state of a fluid supply network 1 based on pressure patterns analysis according to the present invention. The system 28 as it was described above on the FIG 1 comprises a plurality of pressure sensors 6 wherein each pressure sensor 6 is adapted to be located on a pipe 3, 4 of the network of pipes, Each pressure sensor 6 is adapted to make a measurement of fluid pressure in the pipe 3, 4 where the pressure sensor 6 is installed. The system 28 also comprises a monitoring unit 29 configured to perform pressure pattern analysis in accordance with a method of any one of the claims 1 to 5.

In the enhanced embodiment of the system 28 shown on FIG 9 the system 28 additionally comprises at least one pumping station pressure sensor 7 that is adapted to make a measurement of pressure of the fluid provided by the supply network pressurizing component 2 to the network of pipes. The pumping station pressure sensor 7 is installed downstream of the at least one supply network pressurizing component 2 and there is no pipe junctions between the at least one supply network pressurizing component 2 and the pumping pressure sensor 7. Also the monitoring unit 29 of the enhanced embodiment of the system 28 is configured to perform pressure pattern analysis in accordance with a method of the claims 6. While the invention has been illustrated and described in detail with the help of preferred embodiment, the invention is not limited to the disclosed examples. Other variations can be deducted by those skilled in the art without leaving the scope of protection of the claimed invention. Reference numerals

1 - fluid supply network

2 - supply network pressurizing component

3 - main pipe

4 - distribution pipe

5 - consumer

6 - pressure sensor

7 - pumping station pressure sensor

8, 10, 11, 12, 20, 21, 22, 26, 27 - method steps 13 - one historical fluid pressure behavior portrait 17 - statistical deviation

28 - system

29 - monitoring unit