WO2019123475A1

WO2019123475A1 - Low-cost sensor-based system for landslide monitoring and alerts

Info

Publication number: WO2019123475A1
Application number: PCT/IN2018/050217
Authority: WO
Inventors: Varun Dutt; Kapil Agrawal; Shubham Agrawal; Pratik Chaturvedi; Naresh Mali; Venkata Uday Kala
Original assignee: Varun Dutt; Kapil Agrawal; Shubham Agrawal; Pratik Chaturvedi; Naresh Mali; Venkata Uday Kala
Priority date: 2017-12-18
Filing date: 2018-04-16
Publication date: 2019-06-27

Abstract

A low-cost landslide monitoring system with sensing, decision-making, and alert generation components comprising a plurality of different MEMs-based sensors such as temperature and humidity sensor; atmospheric-pressure sensor; light-intensity sensor; rain gauge; accelerometer; soil-moisture sensor; and force sensors for sensing the on-site movement, soil, and weather properties; a master-microcontroller (MMC) unit with a communication network consisting of a Global System for Mobile (GSM) module; a local solid-state storage device; & a Data Management Center (DMC) constituting of Landslide Modeling Software Application (LMSA) and Raw Data Analysis Software (RDAS) to provide intelligent back-end alerting system which generates alerts based on statistical threshold.

Description

TITLE OF THE INVENTION:

LOW-COST SENSOR-BASED SYSTEM FOR LANDSLIDE MONITORING AND ALERTS TECHNICAL FIELD:

Present invention relates to a low-cost landslide monitoring system with sensing, decision making, and alert generation components. More particularly present invention relates to a low-cost micro-electro-mechanical system (MEMS) - based sensor system for local site- specific monitoring of landslides at a minute-scale interval. BACKGROUND ART:

The Himalayan region in India has been prone to rainfall-induced landslides. Changes in the climate are likely to further increase natural hazards like landslides in the hilly regions of north India. During the monsoon season, there is a huge loss of lives and property, transportation services are disrupted, and supply of goods and services is hampered. Early knowledge of an impending landslide can help government agencies and local people to make plan efforts that minimize the adverse impacts. Thus, forecasting of these events at local-scale will improve the preparedness of society in facing landslide disasters. Present invention aims to provide for an inexpensive monitoring system which may help warn people before a landslide disaster occurs. In India, several research organizations are working on using wired and wireless sensors for monitoring various parameters related to landslides. However, the cost of these monitoring systems is presently very high, predictions are not available at a minute-scale, and their predictions are limited to the parameters sensed at only a single site with a few sensors. The prohibitive high costs in these systems are attributable to the use of conventional sensing technologies (e.g., piezometers, wireline extensometers, pore-water pressure sensors). Moreover, these limitations restrict the large-scale deployment of such kind of landslide monitoring systems over a large geographical area. Like in India, various conventional sensing techniques have been used globally also for monitoring landslides. These include visual interpretation of stereoscopic aerial photographs, Satellite Technology, Unmanned Aerial Vehicles (UAVs) -based remote sensing, Digital Elevation Models (DEMs) from airborne laser altimetry data, and Brillouin Optical Time- Domain Reflectometry (BOTDR). Although these sensing techniques provide complementary methods for landslide monitoring and prediction, these techniques are expensive to use. For example, in the city of Portland, where DEMs were used for landslide monitoring, the mapping cost was between USD 400 and USD 600 per square mile. This high cost of technology for landslide monitoring has been a major reason why the city was unable to afford a large-scale DEM implementation. Similarly, high costs and operating constraints associated with UAVs limited their usage to a wider area for landslide monitoring. Recently, a group of researchers deployed a landslide monitoring system that records rainfall, soil- moisture, and landmass movements. However, this system uses costly capacitive sensors, bringing the overall system cost to be close to USD 1000. This system also incurs large drilling costs for deploying the system subsurface. Furthermore, this landslide monitoring system is restricted to sensing the parameters mentioned above and does not incorporate an alert generation unit, which helps warn people about impending landslides. Additionally, researchers have also proposed an integrated landslide-monitoring sensor system using low- cost tilt and position sensors. However, this system employs a limited set of sensors (ignoring weather and soil parameters), and it does not cater to alerting people about impending landslides. Furthermore, some other systems have been proposed recently which use only a limited number of sensors for measuring tilt and moisture, and they do not use other relevant sensors for monitoring the prevailing weather as well as components for alerting people about landslides.

Prior art CN102163363A titled “Landslide real-time monitoring and warning system” discloses a landslide real-time monitoring and warning system. The primary difference is that data are stored at the local-site in their system; whereas, present invention collects and manages data in a remote server enabling real-time monitoring of the landslide. Another major drawback of the said prior art is that the system does not incorporate other parameters like rainfall, force, and soil-moisture. In contrast, present invention takes into account the weather parameters (temperature, humidity, atmospheric pressure, and light intensity) and soil parameters (soil-moisture, force due to the soil, and accelerations) in the said system. Present invention measures the rate of change of these parameters when a landslide occurs. Also, the communication in the prior art is done via wireless transmission modules to the master node. This wireless sensor system could be less reliable and more vulnerable to noise in communication in the field especially when analog signals are to be transmitted. In contrast, present system uses wired transmission modules between sensors and master node wherein the said wired system is more reliable and less vulnerable to noise in communication. Lastly, the prior art does not propose a warning system for issuing alerts while present system proposes an intelligent back-end alerting system which generates alerts based on statistical thresholds. Another limitation of the prior art is that, the intelligence is present in the master node. This makes the system site-dependent or local. However, in present invention, the sensors and master nodes are passive, and intelligence is located on the remote server, which is site-independent or global. Another prior art CN2665723Y titled “Real-time monitoring and alarming device for landslide” discloses a collapse timely monitoring warner which is formed by a monitoring warning center and a plurality of monitoring- warning points. Again, the said prior art does not incorporate other parameters like rainfall, force, and soil-moisture. In contrast, present system adds weather parameters (temperature, humidity, atmospheric pressure, and light intensity) and soil parameters (soil-moisture, force due to soil, and accelerations) & measures the rate of change of these parameters when a landslide occurs. Another major drawback of the prior art system is that it can generate warnings for a range of about 10 kilometers only; whereas, present system is distance independent as it can generate warnings for any distance via SMS. Yet another difference is that the prior art uses GPS devices for measuring location and movement. However, present invention uses accelerometers for precise changes in accelerations at deployment sites.

Yet another prior art CN 101452629 A dated 07-12-2007 titled“Remote monitoring system for disaster of mountain massif coast” discloses a landslide disaster remote monitoring system, and relates to a landslide monitoring system. Again, the said prior art does not propose a warning system for issuing alerts while present invention proposes an intelligent back-end alerting system which generates alerts based on statistical thresholds. Moreover, this prior art also does not incorporate other parameters like force and soil -moisture. Another drawback is that the prior art uses an infrared displacement sensor node, which senses movement or motion via reflection. This method is indirect and depends upon the reflection of infrared rays. However, present invention employs accelerometers for precise changes in accelerations at deployment sites. The accelerometers provide a direct measure of movement due to mechanical oscillators and not only provide linear accelerations but also angular velocities (i.e., rotation). Present system is thus a definite advancement over the existing prior arts. Another non-limiting prior art CN102637351A titled“Landslide monitoring data transmission system based on GSM (global system for mobile communication) network and data analysis method thereof’ provides a landslide monitoring data transmission system based on a GSM (global system for mobile communication) network and a data analysis method thereof. It comprises of a transmitter and a receiver and an upper computer, wherein the transmitter is connected with a sensor, the transmitter communicates with the receiver by virtue of the GSM network, and the receiver is connected with the upper computer. In contrast, present system, uses wired transmission modules between sensors and master node. The master node then sends data via a GSM network to a remote server (the remote server is not local or located on site). In addition, there is local data-logging for backup in present system via a solid-state storage medium. The number of GSM modules in the said prior art system would be much larger compared to present invention (1 GSM module) making it costlier than the present invention. Their invention proposes a monitoring and analyses system. Thus, it does not propose a warning system for issuing alerts. On the other hand, present system proposes an intelligent back-end alerting system which generates alerts based on statistical thresholds. CN102915629A titled“Landslide monitoring device” provides a landslide monitoring device which comprises a transmitter and a receiver. The prior art does not incorporate other parameters like rainfall, force, and soil-moisture & also does not propose a warning system for issuing alerts. An advancement of present invention is that the system collects and manages data in a remote server enabling real-time monitoring of the landslide. Thus, there is no human intervention in the present system (e.g., a worker) and an intelligent algorithm enables generation of warning against impending landslides.

CN103903395 titled“Low-cost landslide early warning recording device based on MEMS accelerometers” discloses a low-cost landslide early warning recording device based on MEMS accelerometers. A major drawback of this prior art is that the system does not incorporate other parameters like rainfall, force, and soil-moisture. Another limitation is that in the said prior art system, a comparative analysis is carried out on the data, and a preset acceleration threshold value is determined. In contrast, in present system, a moving-average z- score movement threshold value is determined in real-time to compute the severity of movement. Moreover, this severity is used to generate different levels of alerts via SMSes. Furthermore, the prior art system uses a general-purpose costly microprocessor; whereas, present invention uses a specific low-cost microcontroller. Moreover, the prior art measures “mountain slope vibration,” which maybe due to landslides as well as other factors (like sound waves or vibrations due to wind). In contrast, present invention uses accelerometers which measure linear accelerations and angular velocities at the deployment sites, where their values are not impacted by other factors like sound waves or vibrations due to the wind.

CN105674945 titled“Seabed landslide monitoring device and method based on MEMS sensor” discloses a seabed landslide monitoring device and method based on an MEMS sensor. Many limitations mentioned in earlier prior arts are also reflected in this prior art. Moreover, the scope of this particular prior art is restricted to seabed landslides; whereas, present invention is applicable to surface landslides.

Another prior art KR101103697 titled“Landslide Warning System Using Rainfall and Soil Moisture Content” also has above mentioned drawbacks which present invention aims to overcome. An Indian patent application 2099/MUM/2014 titled“Flash Flood and Landslide Alerting System” provides specially a system that can early warn about the upcoming flash flood and also landslide for the upstream and downstream areas. This system's topology is entirely different from present invention. The system uses a landslide detecting mechanism consisting of the laser reflected through mirrors and LDR at the receiving side. In contrast, present invention uses a set of sensors including an accelerometer to directly measure linear accelerations and angular velocities. Thus, present invention is mechanical and the prior art system is optical in its working.

Another application 5075/CHE/2013 titled“An Adaptive Energy Management System And Method For Real - Time Landslide Detection” provides an adaptive energy management system and method for real-time landslide detection. The said system uses deep earth probes, whose installation requires boreholes which are costly; whereas, present invention proposes a low-cost system to be installed at the surface. Furthermore, the prior art discloses a system and method of prioritizing the activation of sensors in each of the deep earth probe based on the meteorological, hydrological and soil parameters. In contrast, present invention uses simultaneous measurement of a number of sensors at a constant a sampling rate. Thus, there is no prioritization among sensors and all sensors are equally important. Another difference is that the prior art employs a wireless sensor network consisting of an intelligent wireless probe (IWP) attached to sensors. However, in present invention, the sensors and master nodes are passive and intelligence is located on remote server, which is site-independent or global. Present invention is thus real-time in the sense that communication is done via GSM and decision is taken at the back-end by the intelligent decision-making component. The said prior art proposes the communication via satellite network (VS AT), which could be expensive & thus prohibitive in its wide application at the grass root level. An added advantage of present invention is also that the said system uses renewable energy via a solar panel and uses a low- powered mechanism involving wired sensors.

One more non-limiting prior art is application number 2444/DEL/2013 titled“Landslide Detection and Alerting System Using Wireless Sensor Network”. While present invention uses MEMs technology that makes the system less expensive and achieves the same success at a significantly cheaper cost, the prior art system does not seem to use low-cost MEMS- based sensors. Their system uses a set of geo-technical sensors. Another advantage is that in present invention the system is decentralized, where each wireless sensor node (i.e., microcontrollers + sensors) are standalone units working independently of other wireless sensor nodes while the prior art system appears to be an integration of different wireless sensor nodes.

Another prior art 401/CHE/2011 titled“A Wireless Sensor Network For Monitoring And Detecting Environmental Disasters and Methods Thereof’ relates to Design, Development, and Deployment of a Wireless Sensor Network for Detection of environmental disasters preferably Landslide but uses costly sensors (like pore-pressure transducer and diaelectric moisture sensor, geophone, and strain gauges); thus making the cost of sensing very high. In contrast, present system uses low-cost sensors (like force resistive sensor, resistive soil- moisture sensor, MEMS-based accelerometer, and MEMS-based temperature and humidity sensors) for landslide monitoring and alerting. Another limitation is that the prior art system uses deep earth probes, whose installation requires boreholes which are costly; whereas, present system proposes a low-cost system to be installed at the surface.

Salient features of present invention overcoming the drawbacks & limitations of the prior arts are thus summarized hereunder: -Most of the prior arts do not incorporate other parameters like rainfall, force, and soil- moisture. In contrast, present invention adds weather parameters (temperature, humidity, atmospheric pressure, and light intensity) and soil parameters (soil-moisture, force due to the soil, and accelerations) in the said system. It measures the rate of change of these parameters when a landslide occurs. -Present system discloses an intelligent back-end alerting system which generates alerts based on statistical threshold which has not been disclosed in the above mentioned prior arts.

-The communication in prior arts is primarily done via wireless transmission modules. These wireless sensor systems could be less reliable and more vulnerable to noise in communication in the field especially when analog signals are to be transmitted. To overcome this limitation in the state of the art, present system uses wired transmission modules between sensors and master node. This wired system is more reliable and less vulnerable to noise in communication. -Lastly, the said system uses renewable energy via a solar panel and uses a low-powered mechanism involving wired sensors.

With an aim to fill this technological gap in the patent & non patent state of the art, present invention discloses a robust landslide - monitoring system based on MEMs technology that includes a set of sensors for monitoring weather, movement, and soil properties as well as a communication system for alerting people about impending landslides. The said system does not require drilling and saves the drilling costs involved. Moreover, the system’s portability and light-weight makes it suitable for large-scale deployment at several landslide sites anywhere in the world.

SUMMARY OF THE INVENTION:

Present invention relates to a low-cost landslide monitoring system with sensing, decision making, and alert generation components. More particularly present invention relates to a low-cost micro-electro-mechanical system (MEMS) - based sensor system for local site- specific monitoring of landslides at a minute-scale interval.

The said system uses low-cost MEMs-based sensors and accompanying communication and alerting technologies. The system includes a temperature and humidity sensor, atmospheric- pressure sensor, light-intensity sensor, rain gauge, accelerometer, soil-moisture sensor, and a force sensor for sensing the on-site movement, soil, and weather properties. In addition to these low-cost sensors, the system uses a Global System for Mobile (GSM) service for sending the sensed data at minute-scale intervals to a remote computer. Also, a local solid- state storage device is installed onsite forlocal data storage (in case of a communication failure). Furthermore, it’s uniquely developed algorithm, helps generate alerts of different severity levels from the system for a range of soil movements.

BRIEF DESCRIPTION OF THE DRAWINGS:

FIG 1 is a block diagram of a particular embodiment of the landslide monitoring and warning system with sensors and communication protocol. FIG 2 is an embodiment of the landslide monitoring and warning system showing different sensor modules used in present invention

FIG 3is a lateral view of the landslide monitoring system with sensors deployed at a shallow depth under the slope’s surface. FIG 4 is a flow chart detailing the working of the system from recording data to issuing alerts.

DESCRIPTION OF EMBODIMENTS OF THE INVENTION:

A prototypical system of present invention has been developed and deployed at one of the landslide -prone sites in Himalayan mountain ranges. A typical architectural diagram of present invention [FIG 1] has been envisaged to comprise of different sensors (BS-X and SS- Y), a master-microcontroller (MMC) unit [201], and a communication network [101] supporting landslide event monitoring and alerting. Communications network [101] includes a Global System for Mobile (GSM) [202] module accessible through a SIM card. Network 101 is termed a field network, and it may be one of several such networks deployed in one or more areas of ground that may be prone to landslide activity.

The said field network [101] may also be termed a probe network because of the deployment of plurality of buried probes (BS-X; illustrated using dotted lines in 101) that carry motion, geologic, and hydrologic sensors like accelerometer [208], force-resistive sensors[209], and soil-moisture sensor [210] within a shallow depth (~ 6-inch) from the slope’s surface. Present design envisages a singular rain gauge [203] deployed in the field at the surface that records the rainfall. In present embodiment, a plurality of wired nodes, SS1-SS3, and BS1-BS3 are deployed at the selected site, where each sensor is connected to the MMC [201]. The MMC [201] is responsible for reading the sensor values of both Surface Sensors (SSs) and Buried Sensors (BSs). SSs refer to those sensors that are located on the mud slope’s surface in present invention. These sensors record the weather and light-intensity values. SSs may include temperature and humidity sensor [207], light-intensity sensor [205], atmospheric-pressure sensor [206], and rain gauge [203]. BSs are placed in small cylindrical holes at shallow depth(~ 6-inch)from the surface of the mud-slope. The BSs in present invention measure different soil-related properties. A plurality of sensors including Soil-moisture sensor [210], a force sensor [209], and accelerometer [208] are referred to as BSs in this framework.

The said GSM module [202] receives the sensor data from MMC [201] and sends it to a remote computer located at a remote Data Management Centre (DMC) [FIG l].The GSM module [202] supports several probe networks and uses cellular data to transfer sensor readings in data-packets to the remote DMC.

By way of an example of present embodiment, the said GSM module is built Dual-Band GSM/GPRS engine. It works on frequencies 900/1800 MHz and has an internal TCP/IP stack to connect to the internet via GPRS. It can be used to make calls, send SMS, data-transfer to a remote server and other M2M services. In addition to the GMS module, present invention provides for a local storage (LS) [204] device which is responsible for preserving all the sensor readings in a local memory to be retrieved in case of poor connectivity and inability of GSM modem to log data in the DMC[Network 102].

Furthermore, the DMC [Network 102] includes a Landslide Modelling Software Application (LMSA) and Raw Data Analysis Software(RDAS). The LMSA enables inventors to monitor local rain events to determine the proper thresholds for data captured by different sensors deployed on site. These thresholds when breached would result in a notification, alert, or warning of an impending landslide among registered users. Present invention also provides for a Data analysis software to analyze incoming sensor data in real-time and to determine when the data indicates certain stages of alerts. Moreover, alert services like e-mail and SMS are implemented in the said system to alert people about the probability of landslides.

A definite advantage of present system is that the network architecture illustrated herein is scalable. Plurality of nodes can be deployed on landslide-prone sites within this architecture.Thus present architecture gives scientists and emergency-response personnel the capability of monitoring very large areas of the landslide. Moreover, the spatiotemporal analysis relative to a larger region, as opposed to a local pocket, provides an even better understanding of events that trigger landslides. Network 101 delivers data continuously from a set of both SSs and BSs in what may be a remote mountainous area, where the DMC might be hundreds of miles distant from the monitored area. Therefore, a very lightweight management framework is provided that can handle various network failures, data corruption, packet loss, and congestion problems.

In a desirable embodiment of present invention, BS-l is an accelerometer [208] used to capture the motion of the slope’s surface at a particular location. It is a small chip-like package which has an integrated 6-axis motion tracking that combines a 3-axis gyroscope and a3-axis accelerometer. BS-lcommunicates with the MMC [201] using bus protocol for data transmission and responds whenever the MMC [201] requests an event. In present setting, the BS-l is responsible for detecting x-, y- and z-axis angular velocities on a user-programmable scale of ±250 s, ±500 s, ±l0007s, and ±20007s. Also, BS-l measures accelerations on a user-programmable scale of ±2g, ±4g, ±8g, and ±l6g. Using these three values of acceleration due to gravity, the MMC [201] is programmed to report two more variables called the pitch angle and the roll angle. The notion of pitch and roll angles comes from airways, where pitch means the upwards or downwards movement of the nose of an aircraft about an axis running from wing-to-wing; and, roll means the rotation about an axis running from the nose to the tail of the aircraft. In our invention, we calculate the pitch and roll at a minute-scale interval to understand how many accelerometers deviated from its last position. The said accelerometer [208] is calibrated by aligning each of its axis one-by-one with Earth’s gravitational acceleration (= 9.8m/s ). Thus, when the accelerometer’s z-axis is aligned with the Earth’s gravitational acceleration, then the accelerometer gives -i-lg acceleration along the z-axis and a Og acceleration along the x- and y-axes. Ideally the accelerometer is placed at the landslide side with its z-axis perpendicular to the ground (i.e., the z-axis measures -i-lg acceleration).

In the same embodiment, BS-2 is a soil moisture sensor [210] which measures the volumetric water content in the soil. BS-2 is buried horizontally in a shallow depth at the slope’s surface. The BS-2 sensor uses capacitance to measure dielectric permittivity of the surrounding medium around its electrodes. In soil, dielectric permittivity is a function of the water content. The sensor creates a voltage proportional to the dielectric permittivity, and therefore the water content of the soil. The sensor averages the water content over the entire length of the sensor. There is a certain zone of influence concerning the flat surface of the sensor, but it has little or no sensitivity at the extreme edges.

The said sensor [210] has two components: logic and probe. The probe senses moisture via the moisture effect on the sensing material’s dielectric constant. The logic component uses an analog pin to communicate with the MMC [201]. The calibration of the soil moisture sensor is done in two steps: the first step involves collecting samples of the soil where the system is to be deployed and drying it, so that entire moisture in the soil gets evaporated. In the second step, a small volume of soil is taken and a measured volume of water is poured in equal intervals of time. Water is kept pouring in the soil till the saturation of the soil is reached, i.e., a point where the soil is unable to absorb any more moisture. The analog readings of dry soil and completely saturated soil map to 0% and 100% moisture, respectively.

In the same embodiment, BS-3 is a resistive force sensor [209] that detects the force on a flexible circular surface area. The circular surface contains a polymer thick-film that exhibits a decrease in resistance with an increase in force applied to the surface. This sensor also communicates with the MMC [201] via an analog pin. The force sensor’s calibration is done by putting a known weight on the sensing area and ensuring that the reading from the sensor matches this weight.

In the same embodiment, SS-l is a temperature and humidity sensor [207] that has been used for calculating pressure and relative humidity. It uses a digital pin communication for reporting values of temperature and pressure. This sensor consists of a humidity component, a thermistor, and an Integrated Circuit (IC) on the back side of the sensor. The humidity sensing unit has two electrodes with moisture holding substrate between them. As the humidity changes, the conductivity of the substrate changes or the resistance between these electrodes changes. This change in resistance is measured and processed by the IC which makes it ready to be read by the MMC [201]. On the other hand, a thermistor is a variable resistor that changes its resistance with a change of the temperature. This sensor is made by sintering of semi-conductive materials such as ceramics or polymers to provide larger changes in the resistance with just small changes in temperature. As the temperature increases, the resistivity of this unit decreases. SS-l was calibrated against known values of temperature and relative humidity from values reported by weather stations for a known geographic location.

Furthermore, SS-2 is a chip-like light sensor [205] that senses the intensity of light (in lux), which is incident upon it. Light intensity sensor also knows as a photodetector uses a p-n junction that converts light into current. The light sensor has a built-in l6-bitAnalog-Digital converter and thus reports lux units directly. This sensor uses a bus-protocol for transmitting data to the MMC [201]. SS-2 was calibrated by comparing its values to those of another commercially available light sensor.

Finally, SS-3 is used to measure the atmospheric pressure [206]. This sensor connects to the MMC [201] using an analog pin on a bus-protocol, and it uses the altitude and temperature as parameters to calculate the barometric pressure. This sensor was calibrated by comparing its values to those reported by weather stations for a known geographical location.

Yet another example embodiment of present landslide monitoring system [FIG 3] with different hydrologic and geologic sensors deployed in alandslide -prone area includes a set of buried earth probes (BSs) which are placed in a small cavity [301] to detect soil movement wherein the said cavity [301] includes three BSs. In this example embodiment, an accelerometer sensor BS-l [302]is installed in a shallow hole buried four-six inches below the slope. This sensor is placed in the cavity to measure soil-movements that occur in thetopsoil layers. The soil-movements are initiated in the less-permeable deeper layers and emerge at the topsoil layers. The orientation of the said sensor [302] should be such that the acceleration due to gravity should be acting only along one of the sensor’s axes (x, y, or z) and the other two axes should report a close to zero acceleration due to gravity. These values of pitch and roll are used for evaluating the magnitude of soil-movements that occurred in the last minute- scale interval.

In addition to the accelerometer, present embodiment also envisages two more sensors namely, force sensor [303] and soil-moisture sensor are also buried in said cavity [301]. The force sensor [303] (discussed previously as BS-3) is placed inside the hole facing towards the crest of the hill under the assumption that the force of moist soil on the hill above this sensor will be normal to the surface of this sensor. Finally, the soil-moisture sensor [304] (discussed previously as BS-2) is also installed in said cavity [301] and is placed at a shallow depth and normal to the slope’s surface. The soil-moisture sensor uses resistance to measure dielectric permittivity of the surrounding medium. The sensor averages the water content over the entire length of the sensor.

The described sensors [301, 302, and 303] are connected to the MMC [201], which is placed in a plastic container [305] at the slope surface. The said plastic container [305] also contains surface sensors [SS-l, SS-2, and SS-3], the GSM module [202], and a local solid-state storage device [204]. Present invention uses solar power (306) as the primary power source of on site. Thisnot only makes the invention more robust but also stand-alone for deployment in remote areas where electricity is scarce.

The working of the said system is now disclosed in following steps as preferable means to achieve the desired logical results of present Land slide monitoring & alert generation system:

-Firstly, recording of the rainfall, temperature, humidity, atmospheric pressure, light -intensity, soil-moisture, force, and soil-movement using low-cost MEMs-based sensors at the deployment site using in-expensive sensors such as a soil-moisture sensor, temperature and humidity sensor, a force sensor, accelerometer, light-intensity sensor, rain gauge, and atmospheric-pressure sensor;

-Secondly, communication of sensors with a master microcontroller MMC [201], which then sends data via a GSM module [202] to a remotely located computer;

-Thirdly, storage of the data received from the GSM module in a database;

-Furthermore, an algorithm queries the database and checks if the recently sensed motion values have exceeded predefined thresholds;

-Finally, depending upon the value of thresholds breached the system issues SMS alerts of different severity levels to registered users. The severity of alerts issued by the alert-generation unit is a function of the amount of movement recorded by the accelerometer sensor. The accelerometer and rain gauge sensors are programmed in such a way that whenever they receive a movement or rain, these sensors interrupt a microcontroller, which records the sensed values during a minute-scale period. Then, these accelerations and rainfall values are sent to the remote computer via the GSM module at the end of the minute-scale interval.

An aspect of the present invention also envisages the steps involved in monitoring soil- movements and communicating alerts of different severity to registered users in case of an impending landslide [FIG 4] , as described hereunder: In the first step [401] the accelerometer [208] monitors the soil-movement. It works using an if-else condition, i.e., if the accelerometer [208] detects movement; the microcontroller [201] adds these movements (in terms of angular rotations, W_c, W_g, W_z). In case there is no soil movement, nothing is accumulated in the angular rotation variables (W_c, W_g, W_z).

Next [402], the microcontroller [201] monitors the rain gauge [203]. If the rain gauge [203] detects rain, it sends an intermpt signal to the MMC [201]. The MMC [201] responds by adding the accumulated-rain (in inches) in a global variable. Both the steps 1 and 2 do the continuous monitoring of location and do not miss any movements or rain-pouring. These processes can be treated as independent processes which do not incur any interruption.

In addition to the above two independent processes, a third process [403] is created to check the minute-scale time intervals. The process consists of two variables, wherein the first variable is a boolean, which becomes one after every 8 seconds and the second variable is a numeric variable which is incremented whenever the Boolean variable is set to 1. Whenever the numeric variable reaches the critical value, i.e., a minute-level, the algorithm proceeds to the next step. It is important to note here that the third step [403] is also an independent process like the other two steps [401 and 402].

At step [404], the microcontroller [201] records readings from the other sensors, i.e., temperature and humidity [207], pressure [206], light-intensity [205], soil-moisture [210], and force [209]. Furthermore, it calculates the pitch and roll (as described above) using the acceleration due to gravity. These values are then stored in a local memory storage device [204]. The microcontroller [201] then uploads data to the remote server via GSM module [202]. As a final step, the microcontroller resets the rain-variable, angular rotation variables (W_c, W_g, W_z), and the numeric time-variable which sends back the algorithm to the step 401. The above four steps are followed by the microcontroller for recording sensed values at the local site and uploading data to the server. Step 405 is executed in the back-end after the data is inserted in the database. At step 405, the system determines if any of the angular rotations, i.e., W_c, W_g, W_z are non-zero. If no W values are non-zero, then the process is successfully executed, and no alerts are issued. However, in case of non-zero values of W, the system responds by calculating the z-score. The z-score is calculated as follows: first, all the three omega values, i.e., W_c, W_g, W_z are made positive and then added together to create a new variable W_w3i· Then the system calculates the mean and standard deviation of all the previous non-zero W_wai available in the database. Finally, a z-score is calculated using the formula below:

Where W, n_sLclev are mean and standard deviations of all previous non-zero W s in the database respectively.

After calculation of z-score, the next step (406) is to compare z-score with several thresholds depending upon the requirement. In step 406, the z-score can be used to generate alerts based on the magnitude of z-value. A z-value close to 0 indicates that movements are very close to the mean and thus, less severe alerts are issued, whereas a high positive z-value implies movements are away from the mean and strong alerts may be required for alerting people. In case of negative z-value, however, no alerts are required. The system can adopt several warning-levels and is not restricted to only weak and strong alerts. Various constituents & functionalities of the said low-cost Landslide monitoring & alert generation system are now disclosed hereunder to further explain the invention & its various embodiments:

Having described the preferred embodiments, it will also become apparent to a person skilled in the art that various modifications can be made to the system without departing from the scope of the invention as defined in this document.

ADVANTAGES:

Present invention addresses the problem of soil movement due to slope -failures and landslides. The Himalayan region in India has been prone to rainfall-induced landslides. During the monsoon season, there is a huge loss of lives and property, transportation services are disrupted, and supply of goods and services is hampered. These reasons become supporting evidence for mitigating the problem of landslides and come up with an inexpensive monitoring system which helps warn people before a landslide disaster occurs. Early knowledge of an impending landslide can help government agencies and local people to make plan efforts that minimize the adverse impacts. Existing wired and wireless sensor technologies have been used for developing landslide monitoring systems, where these systems use very expensive monitoring devices like piezometers, wireline-extensometers, and pore-water pressure sensors. These systems are not cost-effective, and the high costs make their large-scale deployment a challenge. Specifically to address the cost challenge, present invention offers a low-cost landslide monitoring and warning system for sensing various soil parameters and weather properties.

INDUSTRIAL APPLICABILITY: Present invention discloses a robust landslide - monitoring system based on MEMs technology that includes a set of sensors for monitoring weather, movement, and soil properties as well as a communication system for alerting people about impending landslides. The said system does not require drilling and saves the drilling costs involved. Moreover, the system’s portability and light weight makes it suitable for large-scale deployment at several landslide sites anywhere in the world.

Early knowledge of an impending landslide can help government agencies and local people to make plan efforts that minimize the adverse impacts. Thus, forecasting of these events at local-scale will improve the preparedness of society in facing landslide disasters. Present invention aims to provide for an inexpensive monitoring system which may help warn people before a landslide disaster occurs.

Claims

CLAIMS: We Claim,

Claim 1: A low-cost landslide monitoring system with sensing, decision-making, and alert generation components comprising of :

-a plurality of MEMs-based sensors including temperature and humidity sensor; atmospheric- pressure sensor; light-intensity sensor; rain gauge; accelerometer; soil-moisture sensor; and force sensors for sensing the on-site movement, soil, and weather properties;

-atleast one master-microcontroller (MMC) unit with a communication network consisting of a Global System for Mobile (GSM) module; a local solid-state storage device; &

-atleast one Data Management Center (DMC) constituting of Landslide Modeling Software Application (LMSA) and Raw Data Analysis Software (RDAS) to provide intelligent back- end alerting system to generate alerts based on statistical threshold.

Claim 2: A low-cost landslide monitoring system as claimed in claim 1 wherein the said system uses wired transmission modules between sensors and master node & is powered by renewable solar energy.

Claim 3: A low-cost landslide monitoring system as claimed in claim 1 wherein a plurality of wired nodes, a plurality of Surface Sensors (SSs), and a plurality of Buried Sensors (BSs) are deployed at the selected site, wherein each sensor is connected to atleast one MMC. Claim 4: A process of monitoring soil-movements and communicating alerts of different severity to registered users in case of an impending landslide, wherein the system as claimed in claim 1 follows the steps:

(i) First the accelerometer monitors the soil-movement wherein it works using an“if-else” condition, i.e., if the accelerometer detects movement; the microcontroller adds these movements (in terms of angular rotations, W_c, W_g, W_z) & in case there is no soil movement, nothing is accumulated in the angular rotation variables (W_c, W_g, W_z).

(ii) Secondly, the microcontroller monitors the rain gauge wherein if the rain gauge detect rain, it sends an interrupt signal to the MMC & the MMC responds by adding the accumulated-rain (in inches) in a global variable.

Claim 5: A process of monitoring soil-movements and communicating alerts of different severity as claimed in claim 4 wherein both the steps (i) & (ii) ensure continuous monitoring of location & can be treated as independent processes which do not incur any interruption.

Claim 6: A process of monitoring soil-movements and communicating alerts of different severity as claimed in claim 4 & 5 wherein in addition to the above two independent processes (i) & (ii), a third process is created to check the minute-scale time interval, wherein:

(iii) The process consists of two variables, wherein the first variable is a boolean, which becomes one after every 8 seconds and the second variable is a numeric variable which is incremented whenever the Boolean variable is set to 1. Whenever the numeric variable reaches the critical value, i.e.,a minute-level, the algorithm proceeds to the next step.

Claim 7: A process of monitoring soil-movements and communicating alerts of different severity as claimed in claim 4, 5, and 6 where in as the next logical step to (i), (ii), (iii),: (iv) the said microcontroller (MMC) records readings from the other sensors on parameters such as temperature and humidity, pressure, light-intensity, soil-moisture, and force ; calculates the pitch and roll using the acceleration due to gravity; stores these values in a local memory storage device ; uploads data to the remote server via GSM module ; and finally the said MMC resets the rain-variable, angular rotation variables (W_c, W_g, W_z), and the numeric time-variable which sends back the algorithm to the steps enumerated in claim 4.

Claim 8: A process of generating alerts of different severity based on the movement values and thresholds by the low-cost landslide monitoring system as claimed in claim 1, 4, 5, 6, and 7 wherein: (a) the said system determines if any of the angular rotations, i.e., W_c, W_g, W_z are non zero.

(b) If no W values are non-zero, then the process is successfully executed, and no alerts are issued.

(c) In case of non-zero values of W, the system responds by calculating the z-score as follows: first, all the three omega values, i.e., W_c, W_g, W_z are made positive and then added together to create a new variable W_M3i· Then the system calculates the mean and standard deviation of all the previous non-zero W_£o£ai available in the database. Finally, a z-score is calculated using the formula below:

where W, n_{s lev} are mean and standard deviations of all previous non-zero Os in the database respectively.

Claim 9: The process of generating alerts of different severity levels based on the movement values and thresholds by the low-cost landslide monitoring system as claimed in claim 1 & 6 wherein after calculation of z-score, the next step is to compare z-score with multiple thresholds depending upon the magnitude of z-value; wherein the said z-values can be interpreted as the magnitude of the movements occurring at the landslide-area & the system can be programmed to generate multiple warning-levels on the basis of z-value such that:

-if a z-score calculated is less than the minimum threshold, then no alerts are issued OR -if the z-score lies between the first threshold level and the second threshold level, then a level- 1 warning alert is issued.

Claim 10: A low-cost landslide monitoring system with sensing, decision-making, and alert generation components as claimed in claim 1 wherein the said DMC [Network 102] includes a Landslide Modelling Software Application (LMSA) and Raw Data Analysis Software (RDAS) so as:

- the said Landslide Modelling Software Application (LMSA) helps monitor local rain events to determine the proper thresholds for data captured by different sensors deployed on site such that these thresholds when breached would result in a notification, alert, or warning of an impending landslide among registered users; and

- the said Raw Data analysis software (RDAS) helps analyze incoming sensor data in real time and to determine when the data indicates certain stages of alerts;

wherein alert services like e-mail and SMS are implemented in the said system to alert people about the probability of landslides.