Method, controller and waste thermal treatment system

Abstract

A waste thermal treatment system (Fig. 1) is controlled by receiving measurement data 601 from a sensor (120, 122, Fig. 1) characterising a property of the waste. Predictive modelling 602 is applied t

Description

The present invention is directed towards a method, controller, and waste thermal treatment system that incinerates waste to generate a combustion gas. The waste thermal treatment system may be a waste-to-energy conversion system that recovers heat from the combustion gas to generate useable energy. BACKGROUND Waste-to-energy conversion systems typically comprises a combustion system that comprises a primary combustion chamber that combusts waste to produce ash and exhausts a combustion gas. A secondary combustion chamber fires the combustion gas to remove pollutants in the combustion gas. The combustion gas flows to a heat exchanger that recovers heat from the combustion gas for use by an energy generator to generate electricity. The now-cooled combustion gas flows through an emission treatment system and is exhausted to the atmosphere. It is known to incorporate sensors within the waste-to-energy conversion system to monitor the operating conditions of the waste-to-energy conversion system in realtime. A controller of the waste-to-energy system can control the conditions of the waste-to-energy conversion system such as the temperature of the primary combustion chamber based on the measurements from the sensors. This is performed to ensure that the waste-to-energy conversion system stays within a set of operating parameters. It is an objective of the present invention to provide an improved mechanism for controlling the operation of a waste thermal treatment system such as a waste-to-energy conversion system. SUMMARY According to the present invention, there is provided a method, controller, and waste thermal treatment system as set out in the accompanying claims. Other aspects of the disclosure are apparent from the dependant claims and the description which follows. According to a first aspect of the disclosure, there is provided a method for controlling a waste thermal treatment system. The method comprises receiving measurement data from a sensor, the measurement data characterising one or more properties of waste to be loaded into a combustion system of the waste thermal treatment system. The method comprises applying predictive modelling to the measurement data to generate a prediction of how the one or more properties of the waste will change the performance of the waste thermal treatment system. The method comprises controlling the waste thermal treatment system based on the generated prediction. The method comprises receiving measurement data from a plurality of sensors, the measurement data characterising the performance of the waste thermal treatment system after the waste is loaded into the combustion system. The method comprises using the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system. Advantageously, the method measures properties of the waste prior to loading into the combustion system and uses the measured properties of the waste to predict how the waste will change the performance of the waste thermal treatment system. This may indicate, for example, how the efficiency of the waste thermal treatment system will be affected by the waste to be loaded into the combustion system. The method proactively controls the operation of the waste thermal treatment system based on the generated prediction to help ensure that the waste thermal treatment system is operating efficiently. Furthermore, properties of the waste thermal treatment system are measured after the waste is loaded into the combustion system to characterise the performance of the waste thermal treatment system in real-time. This data is used to adaptively control the waste thermal treatment system to again help ensure that the waste thermal treatment system is operating efficiently. In this way, the method uses sensor data to achieve efficient control of the waste thermal treatment system such as by maximising uptime, ensuring efficient combustion occurs, ensuring environmental compliance, and ensuring efficient energy harvesting occurs. The predictive modelling may comprise using the measurement data and historical data relating to the performance of the waste to generate the prediction. The predictive modelling may comprise using the measurement data and data indicating the current operational conditions of the waste thermal treatment system to generate the prediction. The data indicating the current operational conditions may be obtained from the plurality of sensors. The measurement data from the sensor characterising one or more properties of the waste may comprise measurement data characterising the type of the waste. Advantageously, the method predicts how the type of waste (e.g., whether the waste is hazardous or non-hazardous) will affect the operating conditions of the waste thermal treatment system and pre-emptively controls the operation of the waste thermal treatment system based on the generated prediction. The measurement data from the sensor characterising one or more properties of the waste may comprise measurement data characterising the weight of the waste. Advantageously, the method predicts how the weight of the waste will affect the operating conditions of the waste thermal treatment system and pre-emptively controls the operation of the waste thermal treatment system based on the generated prediction. The method may comprise receiving measurement data from a plurality of sensors characterising one or more properties of waste to be loaded into a combustion system of the waste thermal treatment system. Controlling the waste thermal treatment system based on the generated prediction may comprise controlling one or more actuators of the combustion system. Advantageously, the method controls one or more actuators of the combustion system based on the prediction generated from the measurement data characterising one or more properties of the waste. This allows the combustion system to be set to operate at the desired operating conditions for the waste to be loaded into the combustion system. Controlling the one or more actuators of the combustion system may change the operational temperature and I or pressure of the combustion system. The one or more actuators of the combustion system may comprise one or more of a fuel injector, an air flow regulator, and a movable grate. Controlling the waste thermal treatment system based on the generated prediction may comprise controlling one or more actuators of an emission control system arranged to treat combustion gas received from the combustion system. Advantageously, the method controls one or more actuators of the emission control system based on the prediction generated from the measurement data characterising one or more properties of the waste. This allows the emission control system to be set to operate at the desired operating conditions for the waste to be loaded into the combustion system. Using the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system may comprise adaptively controlling the rate at which waste is loaded into the waste thermal treatment system. Advantageously, the method adaptively controls the rate at which waste is loaded into the combustion system so as to, for example, maintain a desired combustion efficiency or energy generation rate. Using the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system may comprise adaptively controlling the operational conditions of the combustion system. Using the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system may comprise controlling one or more actuators of the combustion system to adjust the operational temperature and / or pressure of the combustion system. Using the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system may comprises adaptively controlling a rate at which waste and air are delivered to the combustion system. Advantageously, the fuel (waste) and air delivery rates to the combustion system can be adaptively controlled based on sensor feedback received in real-time to ensure that the combustion system is operating at a desired efficiency level. Using the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system may comprise adaptively controlling the operational conditions of an emission control system arranged to treat combustion gas received from the combustion system. Adaptively controlling the operational conditions of the emission control system may comprise controlling one or more actuators of the emission control system, the one or more actuators comprising one or more reagent injectors arranged to inject reagents into the combustion gas. Advantageously, the method may comprises controlling the type of reagent, amount of reagent, and / or rate at which reagent is injected into the combustion gas based on measurement data obtained from the sensors. This can help the system react to changes in the type I level of pollutants present in the combustion gas to help ensure they are removed before the combustion gas is exhausted to the atmosphere. Controlling one or more actuators of the emission control system may comprise controlling a dry sorbent injector to inject dry sorbent reagents into the combustion gas. Controlling one or more actuators of the emission control system may comprise controlling a urea injector to inject urea into the combustion gas. The measurement data from the plurality of sensors may comprise measurement data indicative of the operational conditions of the combustion system. The measurement data from the plurality of sensors may comprise measurement data indicative of the combustion efficiency of the combustion sensor. This measurement data may be obtained from sensors measuring properties at a variety of locations in the waste thermal treatment system such as the combustion system, an emission control system, and, if the system is a waste-to-energy conversion system, a heat exchanger system and an energy generator system. The measurement data from the plurality of sensors may comprise measurement data indicative of the emissions present in combustion gas generated by the combustion system. The measurement data may be indicative of one or more of the type and quantity of emissions present in combustion gas generated by the combustion system. Advantageously, the method may adaptively control the waste thermal treatment system based on the emissions present in the combustion gas to help ensure that emissions are removed from the combustion gas prior to exhausting to the atmosphere. The measurement data may be indicative of the emissions present in the combustion gas before treatment by an emissions control system of the waste thermal treatment system, and after treatment by the emissions control system of the waste thermal treatment system. The measurement data may be indicative of a demand for energy generated by the waste thermal treatment system. Using the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system may comprise implementing one or more adaptive control algorithms. The adaptive control algorithms may use historical data and the measurement data to set a control strategy for the waste thermal treatment system. Advantageously, the adaptive control can improve overtime based on trends identified from the historical data. The one or more adaptive control algorithms may comprise one or more automated feedback loops that correct for deviations from desired operating conditions in realtime. Significantly, the feedback loops may not only correct for deviations, but can also be used to perform diagnostics on the waste thermal treatment system to identify and troubleshoot potential issues before they escalate. The one or more adaptive control algorithms may comprise one or more reinforcement learning algorithms that continuously adapt control strategies based on real-time feedback (from the measurement data) and historical data. The method may further comprises using the measurement data as an input to one or more algorithms to predict potential faults and / or maintenance needs for the waste thermal treatment system. Advantageously, the method uses algorithms such as supervised and / or unsupervised machine-learning algorithms to predict potential faults and / or maintenance needs to thereby reduce downtime and operational costs. According to a second aspect of the disclosure, there is provided a controller for a waste thermal treatment system, the controller comprising: at least one processor; at least one memory; and a communication interface arranged to communicate with a plurality of sensors and actuators of the waste thermal treatment system. The processor is operable to read instructions from the memory and, responsive to the instructions, perform operations comprising: receiving measurement data from a sensor, the measurement data characterising one or more properties of waste to be loaded into a combustion system of the waste thermal treatment system; applying predictive modelling to the measurement data to generate a prediction of how the one or more properties of the waste will change the performance of the waste thermal treatment system; controlling the waste thermal treatment system based on the generated prediction; receiving measurement data from a plurality of sensors, the measurement data characterising the performance of the waste thermal treatment system after the waste is loaded into the combustion system; using the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system. The controller may implement any or all of the features of the method of the first aspect of the disclosure. According to a third aspect of the disclosure, there is provided a control system for a waste thermal treatment system. The control system comprises a sensor associated with the waste input system and arranged to measure one or more properties of waste to be loaded to the combustion system. The control system comprises a plurality of sensors operable to measure conditions in the combustion system and the emission control system. The control system comprises the controller of the second aspect of the disclosure. The control system may further comprise one or more actuators controllable by the controller to control the operating conditions of the waste thermal treatment system. According to a fourth aspect of the disclosure, there is provided a waste thermal treatment system. The system comprises a waste input system. The system comprises a combustion system arranged to receive waste from the waste input system and combust the waste to generate a combustion gas. The system comprises an emission control system arranged to treat combustion gas received from the combustion system. The system comprises a control system. The control system comprises a sensor associated with the waste input system and arranged to measure one or more properties of waste to be loaded to the combustion system. The control system comprises a plurality of sensors operable to measure the performance of the waste thermal treatment system. The control system comprises a controller operable to: receive measurement data from the sensor associated with the waste input system; apply predictive modelling to the measurement data to generate a prediction of how the one or more properties of the waste will change the performance of the waste thermal treatment system; control the waste thermal treatment system based on the generated prediction; receive measurement data from the plurality of sensors; and use the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system. The controller may implement any or all of the features of the method of the first aspect of the disclosure. The combustion system may comprise a primary combustion chamber and a secondary combustion chamber arranged downstream of the primary combustion chamber. The waste thermal treatment system may further comprise a heat exchanger system arranged to recover heat from the combustion gas; and an energy generator arranged to use the recovered heat. BRIEF DESCRIPTION OF DRAWINGS Figure 1 shows a schematic diagram of a waste thermal treatment system according to aspects of the present disclosure. Figure 2 shows a schematic diagram of an example waste input system of a waste thermal treatment system according to aspects of the present disclosure. Figure 3 shows a schematic diagram of an example combustion system of a waste thermal treatment system according to aspects of the present disclosure. Figure 4 shows a schematic diagram of an example heat exchanger system and energy generator system of a waste thermal treatment system according to aspects of the present disclosure. Figure 5 shows a schematic diagram of an example emission treatment system of a waste thermal treatment system according to aspects of the present disclosure. Figure 6 shows a flow diagram for an example method according to aspects of the present disclosure. Figure 7 shows a schematic diagram for an example controller according to aspects of the present disclosure. DETAILED DESCRIPTION Referring to Figures 1 to 5, there is shown an example waste thermal treatment system 100 according to aspects of the present disclosure. The waste thermal treatment system 100 is for incinerating waste materials to produce a combustion gas. The heat from the combustion gas may be recovered to generate useable energy. The waste thermal treatment system 100 (Figure 1) comprises a waste input system 102, combustion system 104, heat exchanger system 106, energy generator system 108, emission control system 110 and chimney stack 112. The waste thermal treatment system 100 further comprises a controller 114 that is operatively connected to sensors and actuators disposed throughout the waste thermal treatment system 100 to control the operation of the waste thermal treatment system 100. The waste input system 102 receives waste to be incinerated, processes the waste and loads the waste into the combustion system 104. The waste input system 102 comprises a waste processor 116 (Figure 2) that shreds and homogenizes the waste to help ensure uniform composition. The waste input system 102 comprises a waste loader 118 that loads the processed weight into the combustion system 104. Sensors 120, 122 associated with (e.g., disposed within) the waste input system 102 monitor one or more features of the waste. The one or more features identify one or more properties of the waste. The sensors 120, 122 comprise a scanner 120, such as a barcode scanner, which obtains information about the waste by scanning a machine-readable code associated with the waste. The machine-readable code may be provided on a bag that the waste is contained in. The information typically identifies the waste type such as by categorizing whether the waste is hazardous or non-hazardous. The scanner 120 typically scans the waste prior to processing by the waste processor 116. The sensors 120, 122 comprise a weight sensor 122 that weighs the waste. It will be appreciated that many different types of waste may be input to the waste thermal treatment system 100. Some examples include agricultural fallen stock, general waste, medical waste, refuse derived fuel, solid recovered fuel, hazardous waste, and aquatic waste. The scanner 120 may identify which type from the above, or other, examples, the waste belongs to. The sensors 120, 122 are operatively connected to the controller 114 and provide measurement data to the controller 114. The controller 114 is operable to control one or more properties of the waste input system 102 such as the waste processor 116 and the waste loader 118. In an example, the controller 114 is operable to control the frequency at which the waste loader 118 loads waste into the combustion system 104. The combustion system 104 receives processed waste from the waste input system 102 and combusts the waste to produce a combustion gas and other waste products such as ash. The combustion system comprises a primary combustion chamber 124 (Figure 3) and a secondary combustion chamber 126. The primary combustion chamber 124 receives waste from the waste input system 124 and combusts the waste to produce a combustion gas. The combustion gas flows to the secondary combustion chamber 126. The secondary combustion chamber 126 operates at a higher temperature than the primary combustion chamber 124 and further combusts the combustion gas to remove harmful components from the combustion gas. Sensors 128, 130 are arranged to measure properties of both the primary combustion chamber 124 and the secondary combustion chamber 126. The sensors 128, 130 are operatively connected to the controller 114 and provide measurement data to the controller 114. The sensors 128, 130 comprise one or a combination of temperature sensors, pressure sensors, draft sensors, and oxygen sensors. The temperature sensors may comprises thermocouples and / or infrared sensors to measure combustion temperature and / or temperature gradients within the primary combustion chamber 124 and / or second combustion chamber 126. The pressure sensors may comprise piezoelectric and / or capacitive sensors to measure pressure within the primary combustion chamber 124, and / or second combustion chamber 126, and / or associated ductwork. The controller 114 is operable to control one or more properties of the combustion system 104 such as by controlling one or more actuators of the primary combustion chamber 124 and the secondary combustion chamber 126. The actuators comprise one or a combination of fuel injectors, air flow regulators and movable grates. Fuel injectors regulate the flow of auxiliary fuels into the combustion chamber 124, 126. Air flow regulators control the supply of air into the combustion chamber 124, 126 for combustion. The air flow regulators typically comprise variable speed fans and dampers. Movable grates control the movement and mixing of waste materials within the combustion chamber 124, 126. The primary combustion chamber 124 is connected to a de-ash system 132 that receives ash output from the primary combustion chamber 124. The ash is stored in ash storage 134 where it may be recycled. Sensors 136 are arranged to measure properties of the de-ash system 132. The sensors 136 comprise a blockage sensor for sensing whether the de-ash system 132 blocked with ash. Sensors 138 are arranged to measure properties of the ash storage 134. The sensors 138 comprise level sensors and weight sensors. The sensors 136, 138 are operatively connected to the controller 114 and provide measurement data to the controller 114. The measurement data is indicative of the performance of the waste thermal treatment system 100 and, in particular, the status of the system 100, whether any faults are present, the combustion efficiency, and the ash level vs waste input. The heat exchanger system 106 receives the combustion gas from the secondary combustion chamber 126 and recovers heat from the combustion gas stream to drive the energy generator system 108. This beneficially captures and reuses the heat generated by the combustion system 104 and cools the combustion gas prior to emissions treatment by the emission control system 110. The heat exchanger system 106 comprises a heat exchanger 140 (Figure 4). The heat exchanger 140 comprises a first fluid path that receives a fluid to be heated by combustion gas and a second fluid path via which the combustion gas flows. Within the heat exchanger 140, heat is transferred from the combustion gas to the fluid. The heated fluid (e.g., steam) from the heat exchanger flows to the energy generator system 108. The energy generator system 108 comprises an electrical generator that uses the heated fluid flow to generate electricity. The electrical generator may comprise a turbine and generator. The turbine is driven by the heated fluid (e.g., steam). The fluid output from the energy generator system 108 flows back to the heat exchanger 140 of the heat exchanger system 106. Prior to the fluid flowing back to the heat exchanger 140, additional heat may be recovered from the fluid by using the heat from the fluid to heat the site at which the waste thermal treatment system 100 is located. If heating of the site is not required, the fluid may flow through a condenser to remove heat from the fluid. The cooled fluid may flow through a treatment system in which one or more sensors are located to measure properties of the fluid such as the flow rate, the fluid level, the pH, and the total dissolved solids. Online cleaning of the heat exchanger 140 may be performed using a cleaning system. The cleaning system may receive compressed air from a compressor. Sensors 142 are arranged to measure properties of the heat exchanger 140. The sensors 142 comprises one or more of level sensors, draft sensors, pressure sensors, temperature sensors, and flow sensors. Sensors 144 are arranged to measure properties of the energy generator system 108. The sensors 144 comprises one or more of level sensors, pressure sensors, temperature sensors, flow sensors, and electrical sensors. The sensors 142, 144 are operatively connected to the controller 114 and provide measurement data to the controller 114. The measurement data is indicative of the performance of the heat exchanger system 106 and the performance of the energy generator system 108. The cooled combustion gas passes from the heat exchanger system 106 to the emission control system 110. The emission control system 110 comprises a gas monitor 146 (Figure 5) that measures the pollutants in the cooled combustion gas prior to treatment. The gas monitor 146 may comprise one or a combination of infrared and electrochemical sensors. The gas monitor 146 may monitor pollutants that would be expected to be released from the combustion of waste such as carbon dioxide, carbon monoxide, nitrogen oxides (e.g., nitric oxide and nitrogen dioxide), sulphur oxides, hydrochloric acid, hydrogen fluoride, total organic carbon, and volatile organic compounds. The emission control system 100 comprises a filtering system 148 that filters pollutants from the combustion gas stream. The filtering system 148 typically comprises one or more ceramic catalyst filters. Upstream of the filtering system, reagents are injected into the combustion gas stream by dry reagent injector 150. The reagents may comprise sorbents and / or activated carbon. If desired, a urea injector 152 is also provided to inject urea into the combustion gas stream for NOx treatment of the combustion gas. Sensors 154, 156 are arranged to measure properties of the dry reagent injector 150 and urea injector 152. The sensors 154, 156 comprise one or more of temperature sensors and level sensors. The sensors 154, 156 are operatively connected to the controller 114 and provide measurement data to the controller 114. Sensors 158 are arranged to measure one or more properties of the filtering system 148. The sensors 158 comprises draft sensors, dust sensors, and temperature sensors. The sensors 158 are operatively connected to the controller 114 and provide measurement data to the controller 114. Used reagent I fly ash is collected from the emission control system 110 by discharge auger. The used reagent I fly ash may be stored and recycled. The treated combustion gas flows from the emission gas control system 110 to chimney stack 112 where it is vented to the atmosphere. Sensors 160 measure one or more properties of the chimney stack 112. The sensors 160 comprise one or more of draft sensors, temperature sensors, particulate matter sensors, emission gas monitors, oxygen monitors and dust monitors. The sensors 160 are operatively connected to the controller 114 and provide measurement data to the controller 114. The emission gas monitor may comprise one or a combination of infrared and electrochemical sensors. The emission gas monitor may monitor for pollutants that would be expected to be released from the combustion of waste such as carbon dioxide, carbon monoxide, nitrogen oxides (e.g., nitric oxide and nitrogen dioxide), sulphur oxides, hydrochloric acid, hydrogen fluoride, total organic carbon, and volatile organic compounds. The particulate matter sensors may comprise one or more laser scattering and light absorption sensors to measure particulate matter levels in the exhaust gasses. The waste thermal treatment system 100 therefore comprises a network of sensors 120, 122, 128, 130, 136, 138, 142, 144, 146, 154, 156, 158, 160 disposed throughout the waste thermal treatment system 100 for monitoring, in real-time, properties of the waste thermal treatment system 100 such as the temperature, pressure, and gas composition within the waste thermal treatment system 100. The sensors 120, 122, 128, 130, 136, 138, 142, 144, 146, 154, 156, 158, 160 are operatively connected to controller 114 which processes the sensor data to adjust operating parameters of the waste thermal treatment system 100. The controller 114 is operatively connected to a network of actuators within the waste thermal treatment system 100 and controls the actuators to adjust the performance of the waste thermal treatment system 100 based on the adjusted operating parameters. In addition to the adaptive and real-time control of the waste thermal treatment system 100, the controller 114 uses predictive modelling based on the measured properties of the waste prior to entry to the combustion system 104 to pre-emptively set operating conditions for the waste thermal treatment system 100 based on the characteristics of the waste. In particular, the controller 114 uses predictive modelling to predict how the one or more properties of the waste will change the performance of the waste thermal treatment system 100. The prediction may relate to a change in the efficiency of the waste thermal treatment system 100 (such as the combustion efficiency and the energy generation efficiency) and the amount / type of pollutants generated by the waste thermal treatment system 100. Based on this prediction, the controller 114 controls the waste thermal treatment system 100 such as by controlling the operational conditions of the combustion system (e.g., the temperature, pressure, and / or gas flow rate), and by controlling the operational conditions of the emission control system (e.g., type, amount, and rate of reagent injection). In this way, the controller 114 preemptively adjusts the operating conditions of the waste thermal treatment system 100 to ensure efficient operation of the waste thermal treatment system 100. The controller 114 also uses real-time monitoring of the waste thermal treatment system 100 using the network of sensors 120, 122, 128, 130, 136, 138, 142, 144, 146, 154,156,158, 160 after the waste is loaded into the combustion system 104 to monitor the performance of the waste thermal treatment system 100. The controller 114 deploys adaptive algorithms to adjust the conditions of the waste thermal treatment system 100 to ensure efficiency operation of the waste thermal treatment system 100. This can help maximize heat output, maximize energy generation, and minimise pollutant generation. Significantly, the controller 114 preforms pre-emptive control of the waste thermal treatment system 100 based on predictions of how the incoming waste will affect performance of the waste thermal treatment system 100 and then performs real-time adjustments to help ensure efficient operation of the waste thermal treatment system. The waste thermal treatment system 100 further comprises one or more user interface components (not shown) that can output information about the operation of the waste thermal treatment system 100 and allow a user to control the waste thermal treatment system 100. The user interface components may be part of the waste thermal treatment system 100 such a local control panel and graphical user interface and / or may include one or more remote devices that are in communication with the controller 114 over one or more communication networks. The one or more user interface components allow users to monitor the performance of the system, view alerts and notifications such as if anomalies are identified by the controller or maintenance is deemed to be needed, and input control functions. Referring to Figure 6, there is shown an example method performed by the controller 114 for controlling the waste thermal treatment system 100. In Step 601, the controller 114 receives measurement data from one or more sensors 120, 122 characterising one or more properties of waste to be loaded into the combustion system 104. The measurement data may include, for example, a code read by the scanner 120 and / or the weight of the waste as measured by weight sensor 122. In Step 602, the controller 114 applies predictive modelling to the measurement data to generate a prediction of how the one or more properties of the waste will change the performance of the waste thermal treatment system 100. The predictive modelling may use additional measurement data obtained from other sensors in the waste thermal treatment system 100 that characterises, for example, the current operational conditions of the waste thermal treatment system 100 such as the current temperature and / or pressure in the combustion system 104. The predictive modelling generates a prediction of how this waste type and waste weight will change the performance of the waste thermal treatment system 100 based on the current operating conditions of the waste thermal treatment system 100. For example, the predictive modelling may indicate that the current operating conditions would lead to an inefficient burn of the waste, a low energy generation efficiency, and an undesirable increase in pollutants generated. For example, the predictive modelling may generate a prediction based on whether the waste is identified as hazardous or non-hazardous. The controller 114 using predictive modelling will be understood as including the controller 114 using machine-learning techniques such as regression models and neural networks trained to predict the effect of changes in waste composition (e.g., the type and weight of waste) on system performance. This prediction is then used to set the one or more control parameters for the waste thermal treatment system 100 to optimes the system performance. Here, optimising the system performance can include reducing emissions, maximizing energy efficiency, and minimizing operational downtime. The machine-learning techniques may involve supervised or unsupervised learning techniques or a combination of both supervised and unsupervised learning techniques. Supervised learning involves training a model based on known input and output data to predict future outputs of the waste thermal treatment system 100. Unsupervised learning allows the controller 114 to find hidden patterns within the data which can then be used to generate algorithms that further increase the effectiveness of the system 100. A combination of machine-learning algorithms may be used. The machine-learning algorithms may comprise decision trees and neural networks to improve the accuracy and robustness of energy output predictions. The machine-learning algorithms can also be used to predict equipment failures using the multitude of sensors used throughout the system. In Step 603, the controller controls the waste thermal treatment system 100 based on the generated prediction. For example, the controller 114 may set operating conditions for the waste input system 102 based on the generated prediction if the generated prediction indicates that the current operating conditions for the combustion system 104 are undesirable. This can include controlling the rate at which the waste loader 118 loads waste into the combustion system 104. For example, the controller 114 may set operating conditions for the combustion system 104 based on the generated prediction if the generated prediction indicates that the current operating conditions for the combustion system 104 are undesirable. The controller 114 may control one or more of the actuators of the combustion system 104 such as fuel injectors, airflow regulators, dampers, valves, and movable grates to set the operating conditions for the combustion system 104. Example operating conditions include the temperature and pressure of the combustion system 104 and, in particular, may include different desired temperatures and pressures for both the primary combustion chamber 124 and secondary combustion chamber 126. For example, the generated prediction may indicate that the current operating conditions of the combustion system 104 will lead to an inefficient bum of the waste materials. In response, the controller 114 controls one or more of the actuators of the combustion system 104 to increase the temperature of the combustion system 104. In this way, the controller 114 proactively changes the operating conditions of the combustion system 104. For example, the controller 114 may set operating conditions for the emission control system 110 based on the generated prediction if the generated prediction indicates that the current operating conditions for the combustion system 104 are undesirable. The controller 114 may control one or more actuators of the emission control system 110 such as one or more reagent injectors 150, 152. The controller 114 may set the type of reagent (e.g., dry sorbent, activated carbon, and / or urea), the amount of reagent to be injected and the frequency of injection based on the generated prediction. The waste that was characterised in Step 601 is loaded into the combustion system 104 for combustion as described below. In Step 604, the controller 114 receives measurement data from a plurality of sensors of the waste thermal treatment system 100 after the waste is loaded into the combustion system 104. The measurement data characterises the performance of the waste thermal treatment system 100. The plurality of sensors may comprise one or a combination of sensors 128, 130 measuring the performance of the combustion system 104, sensors 136, 138 measuring the properties of the ash generated by the combustion system 104, sensors 142 measuring the performance of the heat exchanger system 106, sensors 144 measuring the performance of the energy generator system 144, sensors 146 measuring the pollutants in the combustion gas, sensors 154, 156, 158 measuring the performance of the emission control system 110, and sensors 160 measuring the performance of the chimney stack and properties of the gas emitted by the chimney stack 112. In other words, the controller 114 receives measurement data from a network of sensors strategically arranged throughout the waste thermal treatment system 100 to measure properties such as the temperature, pressure, and gascomposition. The measurement data is received repeatedly and in real-time during the operation of the waste thermal treatment system 100. In an example, the measurement data comprises measurement data from sensors 128, 130 measuring the performance of the combustion system 104. Such properties include the temperature and pressure in the combustion system 104. In an example, the measurement data comprises measurement data relating to the ash output from the combustion system 104. This can include whether the de-ash system 132 is blocked, the weight of the ash, and the level of the ash. This measurement data is indicative of the status of the system 100, whether any faults are present, the combustion efficiency, and is also indicative of the ash level vs the waste input. This measurement data indicates whether the combustion system 104 is operating effectively. In an example, the measurement data comprises measurement data from sensors 146, 160 measuring the gas composition of combustion gas generated by the combustion system 100. This measurement data indicates whether the combustion system 104 is operating effectively or whether an undesirable quantity of pollutants have been generated. In an example, the measurement data comprises measurement data from sensors 142, 144 measuring the performance of the heat exchanger system 106 and energy generation system 108. The measurement data indicates whether the system 100 is operating effectively to generate energy at a desired rate. In Step 605, the controller 114 uses the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system 100. The adaptive control responds to real-time changes to the operation of the waste thermal treatment system 100 to optimise the performance of the waste thermal treatment system 100. The controller 114 uses the measurement data to adaptively control one or more actuators of the waste thermal treatment system 100. In some examples, the controller 114 adaptively controls one or more actuators of the waste input system 102 such as one or more actuators of the waste processor 116 that shreds / homogenizes waste, and the waste loader 118 that loads waste into the combustion system 104. In some examples, the controller 114 analyses the measurement data to set a rate at which waste is loaded into the combustion system 104 by waste loader 116 so as to, for example, optimise combustion parameters and ensure efficient burning. In some examples, the controller 114 adaptively controls one or more actuators of the combustion system 104 so as to adjust the combustion conditions within the combustion system 104. This may involve adaptively controlling one or more fuel injectors, airflow regulators, dampers, valves, or movable grates. In some examples, the controller 114 adaptively controls one or more actuators of the combustion system 104 so as to adjust fuel and air input to the combustion system 104 so as to, for example, optimise combustion parameters and ensure efficient burning. In some examples, the controller 114 adaptively controls one or more actuators of the heat exchanger system 106 so as to adjust the conditions for heat exchange within the heat exchanger system 106. In some examples, the controller 114 adaptively controls one or more actuators of the energy generator system 108 In some examples, the controller 114 adaptively controls one or more actuators of the emission control system 110. This may involve adaptively controlling the type, amount, and / or frequency of the injection of reagent by the dry reagent injector 150 or urea by the urea injector 152. This may involve adaptively controlling the amount and pressure of compressed air delivered to the filtering system 148. The adaptive control of the combustion system 104, heat exchanger system 106, energy generator system 108, and emission control system 110 may depend on one or more of measurement data indicative of the conditions of the combustion system 104 (as measured by sensors 128 and 130 disposed in the combustion system), measurement data indicative of the ash generated by the combustion system 104 (as measured by sensors 136, 138 associated with the de-ash system 132 and ash storage), measurement data from the heat exchanger 106, measurement data from the energy generator system 108, measurement data from the gas monitor 146, measurement data from the emission control system 110 (as measured by sensors 154, 156, 158 disposed in the emission control system 110), and measurement data from the chimney stack 112 (as measured by sensors 160 associated with the chimney stack 112). The controller 114 adaptively controlling the waste thermal treatment system 100 may comprise the controller 114 implementing adaptive control algorithms such as, automated feedback loops to correct for deviations in desired operating conditions in real-time, and reinforcement learning algorithms that continuously adapt control strategies based on real-time feedback from the measurement data and historical data. The adaptive control algorithms may comprise machine learning algorithms. One example machine learning algorithm is Model Reference Adaptive Control (MRAC) that can be implemented adjust control parameters so that the system output (e.g., generated energy) follows the desired reference model. Another example machine learning algorithm is Neural Network-Based Adaptive Control which uses neural networks to model system dynamics and / or directly tune control parameters in real time. Another example machine learning algorithm is Extremum Seeking Control which optimises system performance by finding the input control parameters that maximise or minimise particular parameters, such as to maximise energy output or minimise emissions. Referring to Figure 7, there is shown an example controller 114 according to aspects of the present disclose. The controller 114 comprises one or more processors 162, one or more memory units 164, and one or more communication interfaces 166. The one or memory units 164 are arranged to store real-time measurement data, historical data, and machine-learning models. The one or more processors 162 are arranged to read data / instructions from the one or more memory units 164. The one or more processors 162 implement a data pre-processing module that implements algorithms to clean data, normalize data, and extract features from data. The one or more processors 162 implement a predictive modelling module that implements one or more machine-learning models trained to predict the impact of changes in waste composition (and other operational conditions) on system performance. The one or more processors 162 implement an anomaly detection module that implements one or more models to identify abnormal patterns and potential faults in the system 100. In some examples, the controller 114 is a programmable logic controller. The controller 114 may be local to the waste thermal treatment system 100 or may be remote. The controller 114 may comprise a combination of local and remote (e.g., cloud based) devices. Various combinations of optional features have been described herein, and it will be appreciated that described features may be combined in any suitable combination. In particular, the features of any one example embodiment may be combined with features of any other embodiment, as appropriate, except where such combinations are mutually exclusive. Throughout this specification, the term “comprising” or “comprises” means including the component(s) specified but not to the exclusion of the presence of others. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and / or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and / or steps are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent, or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features. The invention is not restricted to the details of the foregoing embodiment(s). The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.

Claims

1. A method for controlling a waste thermal treatment system, the method comprising:receiving measurement data from a sensor, the measurement data characterising one or more properties of waste to be loaded into a combustion system of the waste thermal treatment system;applying predictive modelling to the measurement data to generate a prediction of how the one or more properties of the waste will change the performance of the waste thermal treatment system;controlling the waste thermal treatment system based on the generated prediction;receiving measurement data from a plurality of sensors, the measurement data characterising the performance of the waste thermal treatment system after the waste is loaded into the combustion system;using the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system.

2. A method as claimed in claim 1, wherein the measurement data characterising one or more properties of the waste comprises measurement data characterising the type of the waste.

3. A method as claimed in claim 1 or 2, wherein the measurement data characterising one or more properties of the waste comprises measurement data characterising the weight of the waste.

4. A method as claimed in claim any preceding claim, wherein controlling the waste thermal treatment system based on the generated prediction comprises controlling one or more actuators of the combustion system.

5. A method as claimed in claim 4, wherein controlling the one or more actuators of the combustion system changes the operational temperature and I or pressure of the combustion system.

6. A method as claimed in claim 4 or 5, wherein the one or more actuators of the combustion system comprises one or more of a fuel injector, an air flow regulator, and a movable grate.

7. A method as claimed in any preceding claim, wherein controlling the waste thermal treatment system based on the generated prediction comprises controlling one or more actuators of an emission control system arranged to treat combustion gas received from the combustion system.

8. A method as claimed in any preceding claim, wherein using the measurement data to adaptively control the waste thermal treatment system comprises adaptively controlling the rate at which waste is loaded into the waste thermal treatment system.

9. A method as claimed in any preceding claim, wherein using the measurement data to adaptively control the waste thermal treatment system comprises adaptively controlling the operational conditions of the combustion system.

10. A method as claimed in claim 9, wherein using the measurement data to adaptively control the waste thermal treatment system comprises controlling one or more actuators of the combustion system to adjust the operational temperature and / or pressure of the combustion system.11 .A method as claimed in any preceding claim, wherein using the measurement data to adaptively control the waste thermal treatment system comprises adaptively controlling a rate at which waste and air are delivered to the combustion system.

12. A method as claimed in any preceding claim, wherein using the measurement data to adaptively control the waste thermal treatment system comprises adaptively controlling the operational conditions of an emission control system arranged to treat combustion gas received from the combustion system.

13. A method as claimed in claim 12, wherein adaptively controlling the operational conditions of the emission control system comprises controlling one or more actuators of the emission control system, the one or more actuators comprising one or more reagent injectors arranged to inject reagents into the combustion gas.

14. A method as claimed in claim 13, wherein controlling one or more actuators of the emission control system comprises controlling a dry sorbent injector to inject dry sorbent reagents into the combustion gas.

15. A method as claimed in claim 13 or 14, wherein controlling one or more actuators of the emission control system comprises controlling a urea injector to inject urea into the combustion gas.

16. A method as claimed in any preceding claim, wherein the measurement data from the plurality of sensors comprises measurement data indicative of the operational conditions of the combustion system.

17. A method as claimed in claim 16, wherein the measurement data from the plurality of sensors comprises measurement data indicative of the combustion efficiency of the combustion sensor.

18. A method as claimed in any preceding claim, wherein the measurement data from the plurality of sensors comprises measurement data indicative of the emissions present in combustion gas generated by the combustion system.

19. A method as claimed in claim 18, wherein the measurement data is indicative of one or more of the type and quantity of emissions present in combustion gas generated by the combustion system.

20. A method as claimed in claim 18 or 19, wherein the measurement data is indicative of the emissions present in the combustion gas before treatment by an emissions control system of the waste thermal treatment system, and after treatment by the emissions control system of the waste thermal treatment system.

21. A method as claimed in any preceding claim, wherein using the measurement data to adaptively control the waste thermal treatment system comprises implementing one or more automated feedback loops that correct for deviations from desired operating conditions in real-time.

22. A controller for a waste thermal treatment system, the controller comprising: at least one processor;at least one memory;a communication interface arranged to communicate with a plurality of sensors and actuators of the waste thermal treatment system,wherein the at least one processor is operable to read instructions from the memory and, responsive to the instructions, perform operations comprising:receiving measurement data from a sensor, the measurement data characterising one or more properties of waste to be loaded into a combustion system of the waste thermal treatment system;applying predictive modelling to the measurement data to generate a prediction of how the one or more properties of the waste will change the performance of the waste thermal treatment system;controlling the waste thermal treatment system based on the generated prediction;receiving measurement data from a plurality of sensors, the measurement data characterising the performance of the waste thermal treatment system after the waste is loaded into the combustion system; andusing the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system.

23. A waste thermal treatment system comprising:a waste input system;a combustion system arranged to receive waste from the waste input system and combust the waste to generate a combustion gas;an emission control system arranged to treat combustion gas received from the combustion system; anda control system comprising:a sensor associated with the waste input system and arranged to measure one or more properties of waste to be loaded to the combustion system;a plurality of sensors operable to measure the performance of the waste thermal treatment system;a controller operable to:receive measurement data from the sensor associated with the waste input system;apply predictive modelling to the measurement data to generate a prediction of how the one or more properties of the waste will change the performance of the waste thermal treatment system;control the waste thermal treatment system based on the generated prediction;receive measurement data from the plurality of sensors;use the measurement data from the plurality of sensors to adaptively control the waste thermal treatment system.

24. A waste thermal treatment system as claimed in claim 23, wherein the combustion system comprises a primary combustion chamber and a secondary combustion chamber arranged downstream of the primary combustion chamber.

25. A waste thermal treatment system as claimed in claim 23 or 24, further comprising a heat exchanger system arranged to recover heat from the combustion gas; and an energy generator arranged to use the recovered heat.

Images