DE102024102480B4 - Determining the blockage status of a roof gutter - Google Patents

Abstract

Method for determining the blockage state of a roof gutter (1), comprising the following procedure steps:
S1) in a training phase ensuring a completely unobstructed gutter (1):
S1a) time-sequential recording of a fill level value describing the fill level of the roof gutter (1) and storage of each recorded fill level value, so that a fill level value reference time series results, and
S1b) recording a precipitation value over time, indicating the amount of precipitation per unit time and area at the location of the gutter (1), and storing each recorded precipitation value, so that a precipitation value reference time series is obtained, and
S2) in a monitoring phase following the training phase without ensuring a completely unobstructed gutter (1):
S2a) Recording the fill level value sequentially over time and saving each recorded fill level value, so that a time series of actual fill level values is obtained,
S2b) recording and storing the precipitation value over time, so that a precipitation value time series is obtained, and
S3) Comparing the actual time series of fill levels and the actual time series of precipitation with the reference time series of fill levels and the reference time series of precipitation and determining the state of obstruction of the gutter (1) on the basis of this comparison.

Description

The invention relates to a method for determining a blockage state of a roof gutter.

Gutters are typically located along the roof edge of a building and serve to collect and channel rainwater. A gutter is usually made of weather-resistant materials such as metal or plastic. Its main function is to collect rainwater falling on the roof and safely direct it to downpipes or drains. This allows for controlled water drainage and helps prevent water damage to the building structure, foundations, and masonry. Furthermore, gutters minimize the risk of soil erosion because the water is channeled in a controlled manner. Regular maintenance is important for optimal performance to ensure that the gutter is free of debris such as leaves and twigs, preventing blockages that could impede the flow of rainwater.

The CN 111640284 A This document describes a wireless roof drainage monitoring system and an early warning system for it. The monitoring system comprises a roof rainfall sensor, a gutter water level sensor, a downpipe water flow sensor, a wireless communication module, and a server. The roof rainfall sensor is mounted on the exterior of the roof, the gutter water level sensor is located in a roof drainage gutter, and the downpipe water flow sensor is located on the surface or outlet of a roof gutter downpipe. The roof rainfall sensor, the gutter water level sensor, and the downpipe water flow sensor are all connected to the wireless communication module, which in turn is connected to the server via a wireless network. Data on rainfall, water level, and flow rate can be obtained in real time. This allows for real-time reporting of gutter blockages during a rain event. However, it is often not possible to detect an actual gutter blockage immediately during a rain event. Therefore, an immediate real-time notification of a blocked gutter often leads to false alarms.

The US 2020/0048910 A1 This concerns a monitoring system for roof gutter systems that uses sensors to record various operating conditions. These include water levels, temperature, rainfall, UV radiation, and the degree of gutter soiling. The sensors can be optical, capacitive, acoustic, or electromagnetic. The data is transmitted wirelessly and can be used to predict maintenance needs or to control other building functions. The system can be integrated into both new and existing gutters.

The CN 115168520 A This describes a system and method for the long-term monitoring of wastewater networks based on the recording of fluid levels and rainfall intensity. Sensors detect anomalies at sampling points, for example, by comparison with historical data or threshold values. An early warning system provides indications of malfunctions, such as blockages, and takes into account different operating conditions (sunny day/rainy day). The data are visualized cartographically and analyzed using AI models (e.g., LSTM) to improve the stability of the wastewater network and enable continuous monitoring.

Based on this, the object of the present invention is to provide a reliable and efficient method for determining a blockage state of a roof gutter.

This problem is solved by the subject matter of claim 1. Preferred embodiments are found in the dependent claims.

According to the invention, a method for determining the blockage status of a gutter is provided, comprising the following process steps: (1) in a training phase, ensuring a completely unblocked gutter: successive recording and storage of a fill level value describing the gutter's fill level, resulting in a fill level reference time series; successive recording and storage of a precipitation value indicating the amount of precipitation per unit of time and area at the gutter's location, resulting in a precipitation reference time series; and (2) in a monitoring phase following the training phase, without ensuring a completely unblocked gutter: successive recording and storage of the fill level value, resulting in an actual fill level time series; successive recording and storage of the precipitation value, resulting in a precipitation reference time series. a precipitation value actual time series is obtained, and comparison of the level value actual time series and the precipitation value actual time series with the level value reference time series and the precipitation Value reference time series and determining the blockage status of the gutter based on this comparison.

The invention provides that a certain number of precipitation events must have occurred during the training phase before the monitoring phase can begin. This ensures that the fill level reference time series contains a certain number of possible fill level values for a completely unobstructed gutter, which can then be assigned to corresponding precipitation values for a completely unobstructed gutter in the precipitation value reference time series. The precipitation values are represented as the amount of precipitation per unit time and area at the location of the gutter. The time series thus generally show the fill levels of the gutter or the precipitation values in a specific temporal sequence. Preferably, the training phase periods for the fill level reference time series and the precipitation value reference time series are identical, allowing for direct correlation of the values. In this context, it is preferred that the time series values are recorded successively at 5-minute intervals during a precipitation event and subsequently at 15-minute intervals.

It is further preferred that the fill level values of the fill level reference time series are determined taking into account the dimensions of the gutter, such as its width, length, height, and/or slope. The fill level values can be recorded using a fill level sensor, which, for example, incorporates a water level sensor in a riser pipe. Alternatively, sensors using capacitive pads to measure the fill level can also be used. Considering the gutter dimensions allows for a more accurate assessment of the fill level. Current precipitation values can be obtained using a rain gauge and/or local weather data via GPS location.

The essential aspect of the invention is that the reference time series for fill levels and precipitation recorded during the training phase are compared with the actual time series for fill levels and precipitation during the monitoring phase. The reference time series provide precise knowledge of the gutter's behavior during and shortly after a precipitation event, thus enabling the detection of gutter blockage based on this comparison. If, during a comparable precipitation event, the comparison of the actual level values with the reference time series reveals that the actual level values during and after the event are generally higher than or above a predetermined threshold, then the gutter is considered blocked. The comparable precipitation event is determined by comparing the reference precipitation time series with the actual precipitation time series. Only using the reference time series determined during the training phase is it possible to determine the blockage status for a specific gutter. Advantageously, the accuracy of the blockage determination can be increased as needed through appropriately detailed training phases.

In general, the degree of blockage can be indicated in various ways. However, according to a preferred embodiment of the invention, the degree of blockage is defined as a finite number of states that indicate the degree of blockage in the gutter with regard to the drainage of rainwater. For example, the blockage of a gutter can be measured using four different states: "not blocked," "slightly blocked," "severely blocked," and "completely blocked." This allows for a uniform and reliable determination of whether and when action is required or the gutter should be cleaned immediately after the condition is identified. In this context, according to a further preferred embodiment of the invention, the degree of blockage can also indicate the extent to which the drainage of rainwater from the gutter is impeded. This can then be expressed, for example, as a percentage: 0% = no impediment to 100% = no drainage. The percentage specification allows for a more precise dosage of maintenance or cleaning intervals and gives the user a more accurate insight into the development of blockage conditions.

According to a preferred embodiment of the invention, the method for determining the blockage state of the gutter comprises the following further step: in the training phase, wherein the gutter is brought into various blockage states, the following steps are taken: the fill level values are successively recorded and stored in the fill level reference time series, and the precipitation values are recorded in the precipitation reference time series for the respective blockage state of the gutter. It is provided that the individual blockage states in the gutter are intentionally induced in order to obtain corresponding fill level and precipitation values. ten. Because these blockage conditions are already included in the level value reference time series and in the precipitation value reference time series, the corresponding blockage conditions can be identified with higher accuracy when comparing the actual time series with the reference time series.

According to a further preferred embodiment of the invention, the method for determining the blockage status of the gutter comprises the following additional steps: in the training phase, ensuring that the gutter is completely unblocked: successive recording of at least one meteorological value, wherein the at least one meteorological value is humidity and/or temperature and/or wind direction and/or wind speed, and storage of each recorded meteorological value, resulting in a meteorological value reference time series; and in the monitoring phase, without ensuring that the gutter is completely unblocked: successive recording of the meteorological value, and storage of each recorded meteorological value, resulting in an actual meteorological value time series, and comparison of the actual fill level value series, the actual precipitation value series, and the actual meteorological value series with the fill level value reference time series, the precipitation value reference time series, and the meteorological value reference time series, and determination of the The system determines the blockage status of the gutter based on this comparison. By incorporating meteorological data, which is compared with the fill level readings, the accuracy of blockage detection can be further increased. This also allows for the detection of false alarms, such as blockages caused by snow or similar obstructions. Furthermore, additional meteorological data, such as wind speed, or general meteorological data from a weather service can be included. This allows for a more accurate description of the precipitation event. These values can originate from an external service, the unit itself, or another technical system, such as a smart home weather station.

In principle, the blockage status can be reported at regular intervals. However, according to a preferred embodiment of the invention, a notification is automatically sent during the monitoring phase if a predetermined blockage level is exceeded. This notification can then be received by a maintenance service, e.g., a roofing company, and/or the user, e.g., a homeowner, who can then clear the blockage accordingly. Because the notification is only sent when a predefined blockage level is exceeded, unnecessary maintenance work on the gutter is avoided. The gutter is maintained efficiently.

In principle, the training and monitoring phases can be of any length. However, according to a preferred embodiment of the invention, the acquisition of fill level values is not terminated during the training and monitoring phases until a fill level of zero has been recorded. This allows for the detection of blockages that occur immediately after a rainfall event. For example, if leaves are washed into a downpipe connected to the gutter via a fluid system and the downpipe becomes blocked, leaving small residual amounts of precipitation in the gutter, these are detected and can be reported accordingly.

It is possible to determine the blockage state in various ways. However, according to a preferred embodiment, the determination of the gutter blockage state is carried out using an artificial neural network. Preferably, the neural network is a recurrent neural network. According to a particularly preferred embodiment of the invention, the determination of the gutter blockage state is further carried out using a time series prediction model. This makes it possible to predict future blockages in the form of blockage states with a certain probability and to issue a notification accordingly before a blockage state occurs. It is even more preferred that the time series prediction model is a long short-term memory model.

Long Short-Term Memory (LSTM) is a type of recurrent neural network (RNN) designed to overcome the vanishing gradient problem in conventional recurrent neural networks. Compared to other RNNs and sequence learning methods like Hidden Markov models, LSTMs are less sensitive to the length of gaps between relevant information. They are used in various applications, including handwriting recognition, speech recognition, machine translation, and healthcare. A typical LSTM consists of a cell and three gates: the input gate, the output gate, and the forget gate. These gates control the flow of information into and out of the cell. The forget gate determines which information from the previous state is discarded, while the input gate determines which new information is stored. The output gate, in turn, controls which information from the current state is output. Through these mechanisms, the LSTM network can maintain useful long-term dependencies for predictions in both current and future time steps.

The LSTM model is particularly suitable for creating time series forecasts due to its ability to capture long dependencies and store information over a longer period for processing time series values.

The LSTM model is provided with the reference time series from the training phase as input. The LSTM model then recognizes patterns and correlations in this input, which, together with comparisons to actual time series, indicate a blockage or an impending blockage during the monitoring phase. If the LSTM model is additionally provided with reference time series for fill levels, precipitation, and meteorological values during the training phase—containing fill levels, precipitation, and meteorological data relevant to blockage conditions—it can achieve high accuracy in predicting and determining the blockage status of the gutter. In a further, preferred enhancement, information on obstructions such as photovoltaic systems, dormers, and satellite dishes is incorporated into the Long Short-Term Memory model. Combined with the gutter dimensions, this allows the LSTM model to be personalized to the specific gutter. Determining the blockage status of the gutter can be done periodically and/or on demand. This allows for flexible decisions, as needed, on how often the blockage status should be determined, either taking energy-saving or safety aspects into account.

According to the invention, a non-volatile, computer-readable storage medium with instructions stored on it is also provided, which, when executed on a processor, effect the method described above.

The invention further relates to a system comprising a level sensor, a communication interface connected to the level sensor for signal transmission, and a processing unit. The communication interface is configured to transmit a level value measured by the level sensor to the processing unit, and the processing unit is configured to execute the method for determining the blockage status of the gutter, as described above. Signal transmission can be wireless or wired. The level sensor is magnetically attached to or mounted on the gutter by means of a snap-in device. For communication with the processing unit, the communication interface includes a microcontroller with a transmitter. As already mentioned, precipitation values can be measured locally using a precipitation sensor and/or retrieved via GPS tracking of the gutter. The same applies to meteorological data. The corresponding sensors are also connected to the communication interface for signal transmission. The sensors and the communication interface can be powered by a battery, rechargeable battery, photovoltaic system, or a connection to a household electrical grid.

The invention is described in more detail below with reference to the drawings and preferred embodiments.

• 1 schematisch ein Ablaufdiagramm für ein Verfahren zum Ermitteln eines Verstopfungszustandes einer Dachrinne gemäß einem bevorzugten Ausführungsbeispiel der Erfindung und
• 2 schematisch ein System zum Ermitteln eines Verstopfungszustandes gemäß einem bevorzugten Ausführungsbeispiel der Erfindung.

The drawing shows

• 1 schematically a flowchart for a method for determining a blockage state of a roof gutter according to a preferred embodiment of the invention and
• 2 schematically a system for determining a blockage state according to a preferred embodiment of the invention.

Out of 1 A schematic flowchart for a method for determining the blockage status of a gutter 1 according to a preferred embodiment of the invention is shown. In a first step S1, it is provided that, during a training phase, while ensuring a completely unblocked gutter 1 in step S1a, a fill level value describing the fill level of the gutter 1 is successively recorded and stored in a fill level reference time series. The fill level values are measured using a 2 The visible fill level sensor 2 measures the water level. Fill level sensor 2 has a water level sensor in a riser pipe. The fill level values are linearized taking into account the dimensions of the gutter 1.

In parallel, step S1b involves the sequential recording of a precipitation value, indicating the amount of precipitation per unit time and area at the location of gutter 1, and the storage of each recorded precipitation value in a precipitation value reference time series. Furthermore, step S1d involves the sequential recording of... Meteorological values. These meteorological values include humidity, temperature, wind speed, and wind direction at the location of the gutter. These meteorological values are stored in a meteorological reference time series.

The training phase also includes introducing various blockage conditions into the gutter. The blockage condition indicates, as a percentage, the degree to which the flow of rainwater from gutter 1 is impeded. In step S1c, the fill level and precipitation values are recorded and stored for each gutter blockage condition in the fill level and precipitation reference time series, respectively. This ensures that several precipitation events occur during the training phase.

The reference series of fill levels, precipitation values, and meteorological values are passed as input values to a Long Short-Term Memory (LSTM) model for training. In a monitoring phase following the training phase, step S2, without ensuring that the gutter 1 is completely unobstructed, the trained LSTM model performs the following steps (S2a to S2c): sequentially recording and storing the fill level value, resulting in an actual fill level value time series; sequentially recording and storing the precipitation value, resulting in an actual precipitation value time series; and sequentially recording and storing the meteorological value, resulting in an actual meteorological value time series.

In a third step (S3), the LSTM model is used to compare the actual time series of water level values, precipitation values, and meteorological values with the reference time series of water level values, precipitation values, and meteorological values. Based on this comparison, the LSTM model then determines the blockage status of the gutter. Simultaneously, the LSTM model can predict future blockages based on the time series values. The blockage status of gutter 1, or the prediction of future blockages, is performed at regular intervals. These intervals can be defined by the user.

2 Figure 1 schematically shows a system for determining a blockage state according to a preferred embodiment of the invention. A level sensor 2 is magnetically attached in the gutter 1. The level sensor 2 is connected to a communication interface 3 for transmitting the level value. A precipitation sensor 7 for measuring precipitation, a thermometer 8 for measuring temperature, a hygrometer 9 for measuring relative humidity, and an anemometer 10 for measuring wind direction and speed are also connected to the communication interface 3 for transmitting their respective values. These values are transmitted to a receiving unit 6 of a processing unit 4 via a transmitter 5 connected to the communication interface. Within the processing unit, the blockage state of the gutter can then be determined based on the transmitted values using the LSTM model. The processing unit can be located locally or on an external server. As soon as a blockage of 50% or more is detected, a message is sent to a maintenance service to arrange for the gutter to be cleaned.

Reference symbol list

1

gutter

2

Level sensor

3

Communication interface

4

computing unit

5

Transmitter unit

6

Receiving unit

7

Rainfall sensor

8

thermometer

9

hygrometer

10

Anemometer

Claims (13)

Method for determining the blockage status of a gutter (1), comprising the following procedure steps: S1) in a training phase, ensuring a completely unblocked gutter (1): S1a) successively recording a fill level value describing the fill level of the gutter (1) and storing each recorded fill level value, resulting in a fill level value reference time series, and S1b) successively recording a precipitation value indicating the amount of precipitation per unit time and area at the location of the gutter (1), and storing each recorded precipitation value S2) in a monitoring phase following the training phase without ensuring a completely unobstructed gutter (1): S2a) recording the fill level value and saving each recorded fill level value, resulting in an actual fill level value time series, S2b) recording the precipitation value and saving each recorded precipitation value, resulting in an actual precipitation value time series, and S3) comparing the actual fill level value and the actual precipitation value series with the fill level value reference time series and the precipitation value reference time series and determining the obstruction status of the gutter (1) based on this comparison. Procedure according to Claim 1 , where the blockage state is a state from a finite number of states which provide a measure of the blockage of the gutter (1) with respect to the drainage of rainwater. Procedure according to Claim 1 , where the degree of obstruction indicates the extent to which the outflow of rainwater from the roof gutter (1) is inhibited. Procedure according to one of the Claims 2 or 3 , comprising the following further step: S1c) in the training phase, whereby the gutter (1) is brought into different blockage states: time-sequential recording and storage of the fill level values in the fill level reference time series and the precipitation values in the precipitation reference time series for the respective blockage state of the gutter (1). A method according to any of the preceding claims, comprising the following further steps: in the training phase, ensuring that the gutter (1) is completely unobstructed: S1d) successively recording at least one meteorological value, wherein the at least one meteorological value is humidity and/or temperature and/or wind direction and/or wind speed, and storing each recorded meteorological value, so that a meteorological value reference time series is obtained, and in the monitoring phase, without ensuring that the gutter (1) is completely unobstructed: S2c) successively recording the meteorological value, and storing each recorded meteorological value, so that an actual meteorological value time series is obtained, and S3) comparing the actual fill level value time series, the actual precipitation value time series, and the actual meteorological value time series with the fill level value reference time series, the precipitation value reference time series, and the meteorological value reference time series, and determining the Congestion status of the gutter (1) based on this comparison. Method according to one of the preceding claims, wherein a message is automatically sent during the monitoring phase if a predetermined blockage state is exceeded. Method according to one of the preceding claims, wherein in the training phase and in the monitoring phase the recording of the fill level values is not stopped until a fill level of zero has been recorded. Method according to one of the preceding claims, wherein the determination of the blockage state of the gutter (1) is carried out using an artificial neural network. Method according to one of the preceding claims, wherein the determination of the blockage state of the gutter (1) is carried out using a time series prediction model. Procedure according to Claim 9 , where the time series forecast model is a Long Short-Term Memory model. Method according to one of the preceding claims, wherein the determination of the blockage state of the gutter (1) is carried out periodically and/or on demand. Non-volatile, computer-readable storage medium containing instructions stored on it which, when executed on a processor, effect a method according to one of the preceding claims. System comprising a level sensor (2), a communication interface (3) connected to the level sensor (2) for signal transmission, and a computing unit (4), wherein the communication interface (3) is configured to transmit a level value measured by the level sensor (2) to the computing unit (4), and the computing unit (4) is configured to perform the method for determining the state of obstruction. the gutter (1) after one of the Claims 1 until 11 to execute.