AU2020101841A4 - Design and development of soft data driven sensors for use in a waste-to-energy (wte) industry plant - Google Patents
Design and development of soft data driven sensors for use in a waste-to-energy (wte) industry plant Download PDFInfo
- Publication number
- AU2020101841A4 AU2020101841A4 AU2020101841A AU2020101841A AU2020101841A4 AU 2020101841 A4 AU2020101841 A4 AU 2020101841A4 AU 2020101841 A AU2020101841 A AU 2020101841A AU 2020101841 A AU2020101841 A AU 2020101841A AU 2020101841 A4 AU2020101841 A4 AU 2020101841A4
- Authority
- AU
- Australia
- Prior art keywords
- data
- waste
- energy
- industry
- plant
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23G—CREMATION FURNACES; CONSUMING WASTE PRODUCTS BY COMBUSTION
- F23G5/00—Incineration of waste; Incinerator constructions; Details, accessories or control therefor
- F23G5/44—Details; Accessories
- F23G5/46—Recuperation of heat
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23G—CREMATION FURNACES; CONSUMING WASTE PRODUCTS BY COMBUSTION
- F23G2207/00—Control
- F23G2207/10—Arrangement of sensing devices
- F23G2207/101—Arrangement of sensing devices for temperature
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23G—CREMATION FURNACES; CONSUMING WASTE PRODUCTS BY COMBUSTION
- F23G2900/00—Special features of, or arrangements for incinerators
- F23G2900/55—Controlling; Monitoring or measuring
- F23G2900/55003—Sensing for exhaust gas properties, e.g. O2 content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
DESIGN AND DEVELOPMENT OF SOFT DATA DRIVEN
SENSORS FOR USE IN A WASTE-TO-ENERGY (WTE)
INDUSTRY PLANT
ABSTRACT
The waste to energy industry plant has gained recent importance as it has a great impact on
sustaining the ecosystem. It bums organic substance in waste materials with thermal reaction and
generates energy. But this process in the industry not only creates ash but also during the process
releases flue gas and synthesis gas, which gets mixed with the oxygen in the atmosphere, causing
pollution. This must be controlled by monitoring the process in the industry and using statistical
parameters like flue gas heating value and the temperature of the synthesis gas. The process
analysis starts with the data requirement gathering along with interfaces like Modbus, OPC, etc.,
using protocols HTTP, MQTT, etc. It involves a cloud gateway, connectivity, and edge to connect
to various processes and devices in the industry to compute and store the data. It is then processed
before modeling by statistical methods to filter the data. Then by deploying deep learning with
nonlinear autoregressive network machine learning, the data is modeled, and predictions are made
for dynamic parameters concerning time. The predictions are preserved to be utilized by the control
center to control the process of waste to energy industry plant.
11 P a g e
DATA ACQUISITION
'VISUALIZATIOEN
RELEVANT DATASETS
FILTERING
PRE-PROCES SING
NON LINE AR AUTOREGRE SSIVE.
NE TWORK MAL4CHINE,
LEARNING
JJJJLJ,%LMNG OF PARAMETERS
MODEL MACINTENAIN
Fig. 2 Process flow diagram
2|Page
Description
NON LINE AR AUTOREGRE SSIVE. NE TWORK MAL4CHINE, LEARNING
Fig. 2 Process flow diagram
2|Page
Description
Field of the Invention.
The Field of invention is related to industry 4.0
To sustain the eco-friendly environment on earth, the management of waste to energy turns attention in the recent years. Recently, technology has improved to deliver energy in the form of heat or electricity from municipal solid waste. In turn there is a need to adopt process to prevent the emission of flue gas and synthesis gas from industries, such as carbon dioxide, carbon monoxide, hydrogen, etc. This invention deploys the design and development of soft data driven sensors for the use in a waste to energy industry plant to perform data analytics of heat or temperature generated in the WTE industry plant.
Background of the invention.
World is looking for a clean environment but still it has to balance and maintain the pollution free ecosystem by deploying a process for reusing the waste that has been generated every day. A thermal treatment known as the incineration where it involves the organic substances present in the waste materials to undergo combustion. From this process, heat and electricity is obtained. But during the process a flue gas and synthesis gas will be released as the ash gets collected due to thermal reaction. It may release gases namely carbon monoxide, carbon dioxide, hydrogen, etc. So, there is a need to monitor the heating value of the flue gas and the temperature of the synthesis gas.
Initially, a mathematical model was deployed in estimating the energy and later, though there were many statistical method of estimation involved, it was time consuming and inaccurate.
1 P a g e
The industries look forward a technology to implement internet of things to automize various process. The HOT requires a cloud based environment in controlling the interface of different sections in the industry. It needs a cloud gateway to connect the collection of data to the HOT and also requires the cloud connectivity to link with edge to handle services and other systems.
The collection of the data from different location and sources in industries deploys protocols like Message Queuing Telemetry Transport to ensure correct devices linked with required path for communication. It also requires various interfaces to connect to cloud connectivity like Modbus or any data exchange standard.
Due to the advancement of technology, soft sensors were deployed in measuring the parameters. In the beginning of using the soft sensors, model driven approach was implemented. In this method, sensor hardware deployed to estimate the parameter for observation and it is interfaced with the process.
Later, data driven approach was deployed. It involved an estimation where there is no requirement for the process to be available but it extracts the data trained and modelled by machine learning algorithm.
Especially, in case real time platform that involves the prediction of the critical parameter, deep learning can be deployed in modelling. In deploying the machine learning algorithm, neural network based nonlinear autoregressive network is more suitable when compared to other linear methods since parameter to be examined is a dynamic parameter with respect to time.
The industrial internet of things platforms deployed in waste to energy plant as there are huge volume of data that has to be recorded due to periodic process that has to be carried out everyday. So, the sensors involved should be of low cost and need to be integrated digitally. It needs a distributed cloud computing platform as there are resources and plant located in various locations are to be monitored and recorded in cloud HOT platform, to make sure that modelling is done with correct prediction of data analytics that makes to take serious action that alters the current ecosystem.
2|Page
Objects of the Invention
The main object of the invention is to design and develop a soft data driven sensors for the use in a waste to energy industry plant. Day by day, there are bulks of waste that has been collected in various locations and it has to be taken to thermal treatment. By deploying cloud and edge, industrial internet of things implemented with soft data driven sensors. The heat statistical parameter of flue gas and the temperature of the synthesis gas are examined in the WTE plant to make predictions for dynamic variations by neural network based nonlinear autoregressive network machine learning. For critical parameter measurement deep learning deployed to control the process in industries.
Summary of the Invention
The waste materials are gathered in huge volume in all the locations and it occupies large land area causing pollution to the residential location affecting the ecosystem. Though there were number of recycling process, there is a technology development required to make use of the waste to be converted to usable energy namely heat and electricity. So a thermal treatment at high temperature is performed by combustion to burn into ashes and convert to usable form. But it may release flue gas and synthesis gas as a result of thermal process. There is a need to examine the flue gas heating value and the temperature of the synthesis gas to make predictions that is used to control the process in the industries. It involves a data driven sensors to gather data from different locations and deploys an HOT platform of cloud gateway and the connectivity. It uses interfaces like Modbus or OPC and deploys protocols like HTTP, MQTT, etc to check, analyze and store the datas. A statistical method has to be deployed to process the data before modeling. Any dependencies data if found, is obtained and filtering is done. Then it is modeled applying the deep learning with nonlinear autoregressive network machine learning and tested. Once predictions are obtained, it is preserved to be used for controlling the process.
Detailed Description of the Invention Fig.1 shows the waste to energy plant diagram. The waste materials that are collected from the residence and commercial places are gathered in the waste to energy treatment plant to undergo thermal treatment. The waste materials are organic substances and are subjected to high temperature to convert into usable form of energy. It is done in a combustion process
3|Page known as incineration which involves a thermal reaction of the substances. This would produce ash and flue gas. The flue gas and the synthesis gas are to be treated in a boiler to generate energy along with heat. The flue gas heating value and the temperature of the synthesis gas has to be monitored by deploying HOT platform with protocols and interfaces in cloud and edge to process the data and model the critical parameters with deep learning with nonlinear autoregressive network machine learning. Then they are preserved to be utilized by the control centre to control the process of industrial plant with the predictions.
Fig. 2 shows the process flow diagram for the design and development of soft data driven sensors for the use in a waste to energy industry plant. The waste to energy plant is deployed with HOT platform cloud and edge. The industry has number of devices and processes. They are linked by cloud connectivity. The data is first gathered in the data acquisition process. The interfaces like OPC and Modbus make ensure that correct communication is deployed in the data collection. It uses protocols like HTTP and MQTT. This check for the data that has been collected is sufficient. It is analyzed and stored in data storage in the HOT platform. It is then preprocessed before performing the modeling with machine learning algorithm. It is preprocessed with statistical methods to visualize any pattern, find any dependent or relevant datasets and if found remove the redundant by filtering. Once it is preprocessed, it is deployed with deep learning with nonlinear autoregressive network machine learning to make the machine to model with the training dataset to make predictions. Once the predictions are made, it is validated. If the required data is inefficient it has to reprocess again or more samples can be analysed. Once testing is done, it has to be again stored in the storage space as maintenance to be used as a record for making any further predictions in the future. By making predictions with the modeled parameters like flue gas heat and the synthesis gas temperature, it can be utilized in the control centre to monitor the variation using smart devices and can control the process. This is an efficient way where it does not require the knowledge of the process to be available but it can make predictions by modeling the data with the machine learning algorithms.
4|Page
We Claim: 1. Waste to energy management plant deployed with the industrial internet of things
2. A high-speed optic fiber connection to perform the computing in various software approaches and also for the communication
3. Highly configured computer to carry out protocols like MQTT, HTTP, etc., in data acquisition.
4. Data communication standard like Modbus, OPC, etc., deployed for interfacing data
5. HOT platforms like cloud connectivity, gateway, edge deployed to provide communication and storage.
6. Statistical methods deployed to process filtering.
7. Deep learning with nonlinear autoregressive network machine learning deployed models dynamic parameter in WTE plant.
8. Smart devices with display to monitor the prediction data in the control center of the process in industries.
11 P a g e
DESIGN AND DEVELOPMENT OF SOFT DATA DRIVEN Aug 2020
DRAWINGS 2020101841
Fig. 1 WTE PLANT
1|Page
Fig. 2 Process flow diagram
2|Page
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| AU2020101841A AU2020101841A4 (en) | 2020-08-15 | 2020-08-15 | Design and development of soft data driven sensors for use in a waste-to-energy (wte) industry plant |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| AU2020101841A AU2020101841A4 (en) | 2020-08-15 | 2020-08-15 | Design and development of soft data driven sensors for use in a waste-to-energy (wte) industry plant |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| AU2020101841A4 true AU2020101841A4 (en) | 2020-09-24 |
Family
ID=72513346
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| AU2020101841A Ceased AU2020101841A4 (en) | 2020-08-15 | 2020-08-15 | Design and development of soft data driven sensors for use in a waste-to-energy (wte) industry plant |
Country Status (1)
| Country | Link |
|---|---|
| AU (1) | AU2020101841A4 (en) |
-
2020
- 2020-08-15 AU AU2020101841A patent/AU2020101841A4/en not_active Ceased
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Design of a real-time monitoring system for smoke and dust in thermal power plants based on improved genetic algorithm | |
| Young et al. | Simplicity out of complexity in environmental modelling: Occam's razor revisited | |
| Krawchuk et al. | Burning issues: statistical analyses of global fire data to inform assessments of environmental change | |
| Akhavian et al. | Coupling human activity recognition and wearable sensors for data-driven construction simulation. | |
| Elfaki et al. | Prediction of electrical output power of combined cycle power plant using regression ANN model | |
| NO329798B1 (en) | System and method for empirical ensemble-based virtual sensing of particulate matter | |
| Williamson et al. | Evolving Bayesian emulators for structured chaotic time series, with application to large climate models | |
| Liu et al. | Machine learning-based stochastic subspace identification method for structural modal parameters | |
| Jhee et al. | Time Series Prediction Using Minimally-Structured Neural Networks: An Empirical Testâť– | |
| AU2020101841A4 (en) | Design and development of soft data driven sensors for use in a waste-to-energy (wte) industry plant | |
| Kumar et al. | Latest innovations in the field of condition-based maintenance of rotatory machinery: A review | |
| Bond et al. | A hybrid learning approach to prognostics and health management applied to military ground vehicles using time-series and maintenance event data | |
| CN114881470A (en) | Data-based waste incineration power generation pollutant emission risk diagnosis and early warning method | |
| Cacciabue | Dynamic reliability and human factors for safety assessment of technological systems: a modern science rooted in the origin of mankind | |
| Wang et al. | Interval estimation of urban ozone level and selection of influential factors by employing automatic relevance determination model | |
| Hollenbeck et al. | MOABS/DT: Methane odor abatement simulator with digital twins | |
| Tikhe et al. | Multicity seasonal air quality index forecasting using soft computing techniques | |
| Winkler et al. | Online modelling based on Genetic Programming | |
| Yan et al. | A dynamic data driven application system for wildfire spread simulation | |
| Satyarthi et al. | Adaptation of Digital Twins as a methodology for management and development of Secure Software Systems | |
| CN110430185A (en) | The method and detection device of loophole are executed for sense command | |
| Patil | Prediction an air quality index data using machine learning and deep learning | |
| Pauzi et al. | Performance comparison of two fuzzy based models in predicting carbon dioxide emissions | |
| Saini et al. | Environmental Scrutinizing System based on Soft Computing Technique | |
| Kortsova et al. | Development of decision support in the structure of the information-analytical system of atmospheric air environmental monitoring |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FGI | Letters patent sealed or granted (innovation patent) | ||
| MK22 | Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry |