AI-POWERED ELECTRONIC NOSE

Abstract

The invention relates to an artificial intelligence-supported electronic nose system that provides solutions to problems such as the inadequacy of institutions capable of measuring odor, the difficulties in delivering odor samples of appropriate quality and quantity to accredited institutions located in different cities, the inability to collect odor samples in a timely manner, and the discrepancies between measurements taken by existing accredited laboratories.

Description

-1- TARIFF AI-POWERED ELECTRONIC NOSE Technical Area 5 The present invention applies to sewage systems in regular and irregular solid waste landfills. in systems, in factory chimneys where production takes place, in wastewater treatment plants, Odor as a pollution parameter in air pollution and quality monitoring stations It relates to the devices and systems that perform the measurements. 10 The invention addresses, in particular, the inadequacy of institutions capable of measuring odor, and their location in different cities. in delivering odor samples of appropriate quality and quantity to accredited institutions difficulties include the inability to collect odor samples in a timely manner and the lack of existing accreditation. Artificial intelligence provides solutions to problems such as discrepancies between laboratory measurements. It is related to an intelligent electronic nose system. State of the Art In regular and irregular solid waste landfill sites, sewage systems, and production facilities. factory chimneys, wastewater treatment plants, air pollution and 20 Because odor is an important pollution parameter in quality monitoring stations. Odor measurements are being taken. In existing systems and with the devices used, institutions that can measure odor insufficient quality and quantity for 25 accredited institutions located in different cities difficulties in delivering the odor sample, timely odor samples such as the inability to obtain the data and differences between the measurements of existing accredited laboratories. Problems are being experienced. Odor parameter and odor-causing VOCs (Volatile Organic Compounds), TOC 30 The connection between (Total Organic Carbon) and odor cannot be established. This causes problems in measuring the units. -2- Generally, smell analyses can be performed using the perception of the human nose, however, human a nasal structure with sufficient sensitivity to the nose could not be created, olfactory perception Problems are being encountered in conducting odor analyses because this is not possible. New technical developments are needed to overcome these problems. 5 Document JPH06213845 (A) describes the odor measurement method and device. To obtain highly reliable and stable data regarding measurements, and in addition An odor measurement system that makes it possible to prolong the lifespan of a semiconductor sensor. A different technique developed to obtain the method is described. 10 In conclusion, the inadequacy of institutions capable of measuring odors, and their presence in different cities, are significant issues. in delivering odor samples of appropriate quality and quantity to accredited institutions difficulties include the inability to collect odor samples in a timely manner and the lack of existing accreditation. 15 new solutions that address problems such as discrepancies between laboratory measurements Technical solutions are needed. Explanation of the Purposes of the Discovery Based on the current state of the art, the aim of the invention is to improve existing structures. an AI-powered electronic nose system that eliminates negative aspects. It is the development of. Another aim of the invention is to determine the odor parameter and the odor-causing VOC (Volatile Organic Compounds). The relationship between Organic Compounds (OCC) and TOC (Total Organic Carbon) was revealed. The goal is to develop an artificial intelligence-powered electronic nose system that enables the placement of a nose. Another aim of the invention is to use odor measurements as a pollution parameter for human analysis. an AI-powered electronic nose system that mimics the nose of a specific 30 When an odor is detected on the odor unit, this indicates a dangerous or undesirable situation. the ability to perceive that it is -3- Another purpose of the invention is to treat sewage in wastewater treatment plants (WWTPs). More accurate results in shorter timeframes in systems and facilities with odor problems. The goal is to develop an electronic nose system that can produce results. Another purpose of the invention is to create a gas / odor tank into which air that has passed through the filters is collected, and 5 It involves the creation of many different types of gas sensors. These sensors are mostly made of metal oxide, These are polymer sensors that have voltage or impedance output. These sensors... The outputs obtained are processed by high-precision analog-to-digital, impedance-to-digital converters. In integrated circuits, the data is converted into digital data and then stored as raw data in a data collection unit. It is an eclipse. 10 Another purpose of the invention is to transmit raw data via wired or wireless communication. thanks to this method, a Wi-Fi module connects to a main calculator, a host computer. This is the sending of data. In this main calculation unit; with the data we have previously obtained (TOC, VOCs (gases, odors) will be passed through an artificial neural network that we will train and displayed on the user screen. It is the transmission of meaningful data. The user can see some key aspects of the electronic nose from the same screen. It can also control its functions. Thanks to this, the measured air is obtained with high accuracy. as meaningful olfactory information in a quick and practical way based on prediction It is presented to the user. Another purpose of the invention is to detect odor using the TSE Air Quality-Dynamic Olfactometer. Olfactometry (human-based odor determination) is based on concentration determination standards. The method involves following similar techniques (such as measurement) and transforming it into an electronic system. In order to achieve the aforementioned objectives, it brings significant technical advantages and technical 25 AI-powered electronics that surpass the known state of the art thanks to innovations. The nasal system has been improved. Explaining the Figures Figure 1; Flowchart of an electronic nose system in a representative application of the invention. It is a drawing. -4- Reference Numbers Detailed Explanation of the Discovery Figure 1 shows the flowchart of the artificial intelligence-assisted electronic nose system. Odors from the odor sample (1) are detected by the expert human nose (2) and the odor information (3) sample concentration data set (4) (ppm) training (5) phase 10 Machine learning algorithm (6) is used after training. Machine learning isolating the odor sample (1) through the inlet pump (15) connected to the algorithm (6). The values ​​received from the sensor array (17) in the gas tank (16) are received by the ADC module (20) By processing the sensor data, it is ensured that it is transferred to the digital environment (21). At this stage, the machine learning algorithm (6) detects the smell from the human nose. values ​​(7) and data from the sensor array (17) via the gas tank (16) odor data processed via communication program (8) and Wi-fi module (9) 1 Scent sample 13 User interface 2 Expert human nose 14. Desired odor value data. shown in the values 3 Odor information 15 Inlet pump 4 Sample concentration data sets 16 Gas tanks Training 17 Sensor array 6 Machine learning algorithms 18 Output pumps Odor level 7, Gas output 19 8 Communication programs 20 ADC modules 9 Wi-Fi module 21. Sensor data is processed and converted into digital form. transfer to the environment Odor data transfer 22 Artificial intelligence module 11 Mainframe 12 Calculator programs -5- The transfer (10) is made to the main computer (11). On the main computer (11) through the calculator program (12) the odor value data to the user interface (13) It is ensured that the desired values ​​are displayed (14). For example, 1000 odor units is high. Action is taken accordingly. If the olfactory unit is high, an electronic nose system is used. It gives a warning. 5 Gas outlet (19) via the outlet pump (18) located after the gas tank (16) The gas from the odor sample (1) is sent to the outside environment. ADC module (20) to which the obtained odor data from the sensor array (17) is transferred 10 These are analog-to-digital converters. They typically convert an analog voltage or current to a continuous voltage or current. data obtained from measurements; calculation, data transmission, information processing, storage and It converts these into numerical expressions for use in control systems. The inadequacy of institutions capable of measuring odors, with only 15 accredited facilities located in different cities... difficulties in delivering odor samples of appropriate quality and quantity to institutions, The inability to collect odor samples in a timely manner and the lack of existing accredited laboratories Solutions are provided to problems such as discrepancies in measurements. Expert person Data received from the nose (2) and from the sensor array (17) in the gas tank (16) The data is processed through a machine learning algorithm (6) and the smell unit of the human 20 It is created by mimicking the nose. In this way, it is taken from the human nose. Accuracy rate of healthy data through artificial intelligence and machine learning high level of odor unit information creation and user interface (13) It ensures that the values ​​are displayed as desired. In regular and irregular solid waste landfills (solid waste storage) sites, sewage in systems, in factory chimneys where production takes place, in wastewater treatment plants, Odor as a pollution parameter in air pollution and quality monitoring stations By making measurements with high accuracy and speed, environmental pollution can be prevented. Passage is ensured. 30 -6- A gas tank (16) into which the air passing through the filters is filled and contains many types of gas A sensor array (17) containing a sensor was created. These sensors are mostly metal. Oxide and polymer materials are used to create sensors with voltage or impedance output. The outputs from these sensors are high-precision analog-to-digital and impedance-to-digital. In converter integrated circuits, the raw data is converted into digital data and then processed in a data acquisition unit. It is ensured that it is stored in data form. Raw data is transmitted via a wired or wireless communication program (8) through Wi-Fi The module (9) ensures that it is sent to a main calculator, the main computer (11). The calculator program (12) on the main computer (11); with the data we have acquired beforehand 10 (TOC, VOC, gases, odor) will be passed through an artificial neural network that we will train and presented to the user. It transmits meaningful data to the user screen on his face (13). The user from the same screen It can also control some of the main functions of the electronic nose. This allows for the measurement of... weather, in a fast and practical way, meaningfully provides a highly accurate forecast. It is presented to the user as a scent information (3). 15 The invention relates to an artificial intelligence-powered electronic nose system, the feature of which is its ability to detect smell. the odor sample (1) from which the value will be measured, from the odor sample (1) or from the environment The scent is sent to the expert human nose (2) and the gas tank (16), expert human nose (2) Machine learning, which is trained and sent with recognized odor information (3) 20 The artificial intelligence module (22) working with the algorithm (6) has at least one inside the gas tank (16). sensor array (7), by transferring the odor data received from the sensor array (7) to the digital environment transmitted to the artificial intelligence module (22), the human nose via the artificial intelligence module (22) Odor data obtained by mimicking scent is transmitted via wired or wireless means. The main 25 that enable warnings to be given when high values ​​are displayed on the data. It is characterized by containing a computer (11). The gas inlet pump (15) through which the gas is sent via the mentioned odor sample (1), insulated gas tank (16) through which gas is conveyed via inlet gas pump (15), gas Gas outlet from which odor data is taken by odor sensors from tank (16) 30 Includes pump (18) and gas outlet (19). -7- Human nose's olfactory recognition data and the odor inside the gas tank (16) Odor data received from the sensor array (17) created with sensors is sent Includes artificial intelligence module (22). Trained olfactory data from the human nose and 5 from the gas tank (16) by comparing odor data and digitizing the odor data a calculator program that is sent via wired or wireless communication for calculation. (12) The main display of the odor unit value (13) on the user interface via (12) Includes computer (11). By training the human nose and olfactory sensors to process odor data, the human nose... artificial intelligence that detects olfactory data, like an electronic nose exhibiting such behavior. Digital odor data from the intelligence module (22) and the artificial intelligence module (22) It includes the main computer (11) which displays the odor unit value transmitted. Gas sent from the odor sample (1) passes through filters and the air fills with gas tank (16), gas tank (16) containing many types of gas sensors metal oxide, polymer Sensor arrays arranged as sensors with voltage or impedance output (17), Outputs from sensors are high-precision analog-to-digital and impedance-to-digital. In converter integrated circuits, the raw data is converted into digital data and stored in a data acquisition unit. It includes a database where data is stored. Raw data from the artificial intelligence module (22) via wired or wireless communication The program (8) and the Wi-Fi module (9) are sent to a main calculator, the calculator program (12); trained with the data we have acquired beforehand (TOC, VOC, gases, odor) 25 passed through the artificial neural network and meaningfully displayed on the user screen (13) of the user interface It includes the host computer (11) to which the odor information (3) or odor value is transmitted as data. The invention is an artificial intelligence-powered electronic nose working algorithm, whose feature is its ability to detect smell. Odors taken from the odor sample (1) whose value will be measured are sent to the expert human nose 30 (2) and also sending to the isolated gas tank (16), expert human nose (2) Odor information (3) Sample concentration data set (4) (ppm) -8- trained in the training (5) stage, the artificial intelligence module that works with machine learning (22) transfer of odor data from the sensor array (17) in the gas tank (16) to the ADC Processing sensor data through module (20) and transferring it to digital environment (21), Transmission of sensor data to the AI ​​module (22), AI module (22) The transmitted human nose olfactory data and olfactory data received from olfactory sensors, human 5 training data in the AI ​​module (22) to detect smell like the nose data generation, wireless or wired transmission of generated odor data transfer to the main computer (11), the calculator program (12) on the main computer (11) Creating odor values ​​through the user interface (13) The display of values ​​is characterized by including the processing steps. 10 Communication program (8) and Wi-fi module (9) from the artificial intelligence module (22) This includes the process of transferring the odor data (10) to the main computer (11). Gases from the odor sample (1) or from the environment are passed through filters and gas 15 gas is sent from the inlet pump (15) to the insulated gas tank (16), odor sensors with metal oxide, polymer voltage or impedance output in tank (16) After the odor data is received, gas output is via the gas outlet pump (18) (19) The release of gas into the external environment, and the high outputs received from these sensors... precision analog-to-digital, impedance-to-digital converter integrated circuits convert digital data to 20 transformed and stored as raw data in a database in a data collection unit. It includes the process phase. The smell of gases released from the odor sample (1) is detected through the nose of a specialist human (2). 25 during the training (4) phase via the sample concentration data set of knowledge (3). This includes the process of training and sending the data to the artificial intelligence module (22). By comparing the data from the expert human nose (2) and the odor sensors It works with machine learning that enables the matching and generation of scent data. Generating scent data in the AI ​​module (22), 30 from the AI ​​module (22) raw data via wired or wireless communication program (8) and Wi-Fi Sending the module (9) to a main calculator, the calculator program (12) -9- an artificial neural network trained with previously acquired data (TOC, VOC, gases, odor) passed through the network and as meaningful data to the user screen (13) on the user interface This includes the process of transmitting the olfactory information (3) or olfactory value. In the artificial intelligence module (22), odor data is generated and odor data is 5 This includes the process of sending the data to the main computer (11) for calculation.

Claims

-10- REQUESTS 1. The invention relates to an artificial intelligence-assisted electronic nose system, the feature of which is; - the odor sample to be measured (1), - the odor sample (1) or the odor from the environment is sent to the expert 5 human nose (2) and gas tank (16), - trained to recognize smell information (3) through the expert human nose (2) artificial intelligence working with the machine learning algorithm (6) to which it is sent module (22), - at least one sensor array (7) inside the gas tank (16), 10 - by transferring the odor data received from the sensor array (7) to the digital environment to which it is transmitted artificial intelligence module (22), - obtained by imitating the human nose through the artificial intelligence module (22) scent data is transmitted via wired or wireless means.

15. By displaying the data values, a warning is given at high values. It is characterized by containing the main computer (11) that provides it.

2. An artificial intelligence-assisted electronic nose system conforming to Claim 1, whose features include: - gas inlet through which the gas is sent via the mentioned odor sample (1) pump (15), 20 - insulated gas tank through which gas is conveyed via inlet gas pump (15) (16), - gas outlet from which odor data is taken by odor sensors from the gas tank (16) It includes a gas outlet pump (18) and a gas outlet (19).

3. An artificial intelligence-assisted electronic nose system conforming to Claim 1, whose features include: - data of the human nose's ability to recognize smell and of the gas tank (16) taken from the sensor array (17) formed with odor sensors inside It includes an artificial intelligence module (22) to which the smell data is sent. -11- 4. An artificial intelligence-assisted electronic nose system conforming to Claim 1, whose features include: - trained olfactory data coming from the human nose and from the gas tank (16) Comparing the incoming odor data and digitizing the odor data via wired or wireless communication for calculation by transmission sent via the calculator program (12) on the user interface 5 (13) contains the main computer (11) where the value of the odor unit is displayed.

5. An artificial intelligence-assisted electronic nose system conforming to Claim 1, whose features include: - odor data from the human nose and olfactory sensors like an electronic nose that can be trained to behave like a human nose 10 AI module (22) in which odor data is detected and AI Digital odor data from module (22) is transmitted to the odor unit It includes the main computer (11) where the value is shown.

6. An AI-assisted electronic nose system conforming to Claim 1, with the following features: 15 - the gas sent from the odor sample (1) passes through the filters and into the air the gas tank (16), - metal oxide containing several types of gas sensors inside the gas tank (16), polymer sensors designed as voltage or impedance output sensor array (17), 20 - the outputs from these sensors are high-precision analog-digital, Impedance-to-digital converters convert data into digital data in integrated circuits. a database where raw data is stored in a data collection unit It includes.

7. An artificial intelligence-assisted electronic nose system conforming to Claim 1, whose features include: - raw data from the artificial intelligence module (22) wired or wirelessly to a main calculator with communication program (8) and Wi-Fi module (9) sent, calculator program (12); with the data we have acquired beforehand (TOC, VOC, gases, odor) are passed through a trained artificial neural network and 30 the user screen on the user interface (13) uses odor as meaningful data -12- the host computer to which the information (3) or the odor value is transmitted (11) It includes.

8. The invention is an artificial intelligence-assisted electronic nose working algorithm, the feature of which is; - 5 of the odors taken from the odor sample (1) in which the odor value will be measured to the expert human nose (2) and also to the isolated gas tank (16) sending, - sample of olfactory information (3) data from expert human nose (2) trained with concentration data set (4) (ppm) during training phase (5) Transfer to the AI ​​module (22) that works with machine learning, 10 - ADC module of odor data from sensor array (17) in gas tank (16) (20) processing sensor data and transferring it to digital environment (21), - transmitting sensor data to the AI ​​module (22), - Human nose odor data and odor 15 transmitted to the AI ​​module (22) Odor data received from sensors can detect smells, just like the human nose. Training data in the AI ​​module (22) for perception of smell data generation, - generated odor data is transmitted wirelessly or via wired to the main system. transfer to the computer (11), 20 - scent via calculator program (12) on the main computer (11) the creation of values, - display of vagina values ​​(13) on the user interface It is characterized by including stages.

9. This is a working algorithm that complies with claim 8, and its characteristic is; - communication program (8) and Wi-fi module from the artificial intelligence module (22) (9) transfer of odor data to the main computer (11) (10) It includes the process stage. -13- 10. This is a working algorithm that complies with claim 8, and its characteristic is; - odor sample (1) or gases coming from the environment through filters passed through the gas inlet pump (15) to the insulated gas tank (16) sending, - gas tank (16) metal oxide, polymer voltage or impedance output 5 After the odor data is collected by the odor sensors, the gas outlet... gas is released to the outside environment through the pump (18) (19) sending, - the outputs from these sensors are high-precision analog-digital, Impedance-to-digital converters convert data into digital data in integrated circuits. stored in a database as raw data in a data collection unit It includes the process stage.

11. This is a working algorithm that complies with claim 8, and its characteristic is; - gases released from the odor sample (1) detected by a human nose (2) 15 sample concentration data set of odor information (3) obtained from trained through the artificial intelligence module (22) during the training (4) stage The sending process involves a stage of the process.

12. The working algorithm is compliant with claim 8 and its characteristic is; 20 - data from expert human nose (2) and odor sensors This allows for the comparison and matching of scents to generate olfactory data. smell in the artificial intelligence module (22) that works with machine learning data generation, - raw data from the artificial intelligence module (22) via wired or wireless 25 to a main calculator with communication program (8) and Wi-Fi module (9) sending, - the calculator program (12) with the data we have already obtained (TOC, VOC, Gases (odors) are passed through a trained artificial neural network and presented to the user. the smell information (3) 30 as meaningful data on the user screen on his face (13). or the transmission of the odor value involves the processing stage. -14- 13. This is a working algorithm that complies with claim 8, and its characteristic is; - by generating odor data in the artificial intelligence module (22) the process of sending data to the main computer (11) for calculation It includes stages 5.

Images