Pyrolysis plant and operating method thereof

Abstract

The present invention describes a pyrolysis plant for the pyrolysis of industrial and urban organic waste including apparatus such as means for receiving and conditioning waste, pyrolysis reactors, combustion chambers for supplying thermal energy, a rectification tower, a vapour condensation apparatus, a non-condensed gas scrubbing tower, cooling apparatus, and tanks for oil and wastewater. The plant is provided with a control system with sensors to monitor input and output variables, and mechanisms to regulate the operation of the apparatus. It integrates two analysers based on artificial neural networks: a first artificial neural network adjusts the temperature of the combustion chambers, and a second artificial neural network controls the temperature of the rectification towers, working in tandem to optimise the energy efficiency and performance of the process.

Description

[0001] DESCRIPTION

[0002] Pyrolysis Plant and Operating Method Thereof

[0003] TECHNICAL FIELD

[0004] The present invention falls within the field of devices, systems, and methods for obtaining pyrolytic oils from the recovery of industrial and urban plastic waste.

[0005] BACKGROUND OF THE INVENTION

[0006] Pyrolysis plants represent an advanced technological solution for treating and recovering waste, transforming it into high added value products by means of a thermochemical process. Pyrolysis, which takes place in the absence of oxygen, breaks down complex organic materials into simpler fractions, mainly generating three products: combustible gases (such as hydrogen and methane), liquids such as oil, and carbon-rich solid waste, known as char or coke.

[0007] These plants are particularly relevant in the management of plastic waste, used tyres and biomass, which are sectors where traditional disposal methods, such as landfills or incineration, cause significant environmental problems. By reducing reliance on landfills and reducing emissions associated with combustion, pyrolysis is a sustainable and efficient alternative within the circular economy.

[0008] In addition, pyrolysis plants offer economic benefits by recovering materials and generating energy, contributing to a cleaner development model. However, their implementation requires addressing technical, regulatory, and economic challenges to maximise their potential in integrated waste management.

[0009] Today, advanced control systems are essential for the efficient operation of pyrolysis plants, especially when real-time process modelling is required. However, real-time control of the different units that make up the pyrolysis plant presents certain complexities that conventional methods, such as temperature control by PLC (Programmable Logic Controller) systems, cannot handle in an optimal manner due to the nature of the process and the heterogeneity of the input materials. It is in this context that artificial intelligence methods show their ability to overcome the limitations of traditional approaches, offering more adaptive and effective solutions.In this context, artificial neural networks have become increasingly important in recent years in the field of chemical engineering, where they have been widely applied to optimise various processes and improve decision-making.

[0010] Essentially, a neural network consists of a set of artificial neurons or perceptrons connected to each other. An artificial neuron (AN) is a processing unit that attempts to mimic the behaviour of a biological neuron, such as those that make up the human brain. Each neuron receives multiple inputs, which it combines, usually by summation. This summation is adjusted by a transfer function, and the value resulting from this function is sent directly as an output of the processing element.

[0011] In an AN, each input is multiplied by its weight or weighting, similar to the synaptic connection between biological neurons. The weighted inputs are summed and the activation level of the neuron is calculated. The output of the AN is connected to other neurons via weighted connections. In a typical ANN, neurons are organised in layers, with connections between adjacent layers to form a sequential structure.

[0012] There are two layers that connect to the external environment: the input layer, which acts as a buffer where data is presented to the network, and the output layer, which stores the response of the network to that input. The other layers are known as hidden layers.

[0013] In the development of an artificial neural network (ANN), the network learns the rules for processing information by adjusting the weights associated with the connections between the neurons of the different layers that make up the network. The aim of training an ANN is to ensure that, for a specific set of inputs, the network generates a set of outputs that are as close as possible to the expected results or, at least, are consistently acceptable. This training process is based on the sequential application of various input data sets or vectors, adjusting the weights of the interconnections according to a predetermined learning algorithm. During training, the weights are progressively updated, gradually converging to optimal values that allow the network to produce the desired outputs for each input provided.

[0014] Training algorithms for the artificial neural network are divided into two categories: supervised and unsupervised. In supervised training, each input vector is paired with its desired output vector. The process consists of presenting an input vector to the network, calculating its output, comparing it with the expected output, and using the difference (error) to adjust the weights ofthe connections. This adjustment follows an algorithm that seeks to minimise error. The pairs of vectors in the training set are processed sequentially and repetitively. The adjustment of weights continues until the error in the entire training set is sufficiently small and acceptable.

[0015] A key characteristic of artificial neural networks is that they learn their own rules from examples provided to them during the training phase. This learning is achieved by means of a rule that adjusts the weights of the connections according to the input examples and, optionally, the desired outputs.

[0016] Another important aspect of ANNs is how they store information. Their memory is distributed over the weights of all the connections of the network. In addition, some ANNs are “associative”: they can recognise partial inputs, look up the most similar one stored in their memory and generate the output corresponding to the complete input.

[0017] Back-propagation networks are particularly useful in this technical field. These networks use a supervised training method, in which pairs of patterns are presented: an input pattern and its corresponding desired output. With each presentation, the weights are adjusted to reduce the error between the desired output and the response of the network. The back-propagation algorithm includes a forward and a backward propagation phase, which are performed for each pattern during training.

[0018] The forward propagation phase starts when a pattern is present in the input layer. Each input unit takes the corresponding value of the pattern and the output level of the first layer is calculated. The remaining layers then perform forward propagation to determine the activation level of each one. Once this phase is completed, back-correction or propagation begins. Calculations to adjust the weights of the connections start at the output layer and go backwards through all the layers to the input layer. The weight adjustments are divided into two groups: those for the units of the output layer and those for the units of the hidden layers.

[0019] Adjusting the weights in the output layer is simple, as the desired value for each unit is known. Each unit generates a real value, which is compared to the expected value in the pattern of the training set.

[0020] As for the adjustment of the weights of the hidden layers, they do not have a desired output vector, so it is not possible to apply the error propagation method used in the processing units of the output layer. The value of the error calculated for units of this type is obtained frommathematical equations, not detailed in this description, which allow the definition of a variable q, known as the learning coefficient. This coefficient, generally between 0.25 and 0.75, reflects the degree of learning in the network.

[0021] The present invention provides a pyrolysis plant equipped with a control system based on the joint use of artificial neural networks. This system, once trained with the input and output variables of the process, is able to control in real time the temperature in critical units of the plant, such as the combustion chamber that supplies thermal energy to the reactor. This ensures optimised operation of the plant’s resources during operation.

[0022] DESCRIPTION OF THE INVENTION

[0023] The present invention discloses in its first aspect a pyrolysis plant for the pyrolysis of industrial and urban plastic waste, comprising at least the following apparatus: means for receiving and conditioning the waste, at least one pyrolysis reactor, at least one combustion chamber configured to provide thermal energy to the pyrolysis reactor(s), at least one rectification tower, at least one gas condensation apparatus, at least one non-condensed gas scrubbing tower, at least one indirect contact cooling apparatus, at least two tanks configured for receiving light and heavy oil fractions, and at least one tank for receiving wastewater, wherein the plant is provided with a control system characterised in that it comprises:

[0024] - at least one installation with sensors and measuring instruments associated with the apparatus of the plant whereby they detect the values of the input and output variables of said apparatus;

[0025] - at least one installation for regulating or controlling the operation of the apparatus of the plant;

[0026] - a first analyser for analysing the temperature of the combustion chamber(s) (T-06) in the form of a first artificial neural network (ANN1), such that it emits, from the measured values, control quantities to the regulation and control installations of the apparatus of the plant to adjust T-06 to a set temperature;

[0027] - a second analyser for analysing the temperature of the rectification tower(s) (T-01) in the form of a second artificial neural network (ANN2), such that it emits, from the measured values, control quantities to the regulation and control installations of the apparatus of the plant to adjust T-01 to a temperature equal to or less than 315°C;wherein the second analyser is activated to operate in tandem with the first analyser when T-01 exceeds a threshold value (Tu), and wherein the threshold value is 290°C, and wherein the first and second analyser operating in tandem determine T-06 set temperature.

[0028] In the second aspect of the invention, it relates to a pyrolysis process for the pyrolysis of industrial and urban plastic waste by means of a pyrolysis plant according to the first aspect of the invention, which comprises the operational steps of waste treatment, pyrolysis of the waste and condensation of the outlet vapours of the reactor, characterised in that the temperature of the combustion chamber(s) (T-06) and the temperature at the head of the rectification tower(s) (T-01) is controlled by artificial neural networks (ANNs), and said process comprises:

[0029] - acquiring values of the input variables of at least one operational step of the process by means of sensors and measuring instruments;

[0030] - recording values of the temperature T-06 and the temperature T-01;

[0031] - presenting the ANN1 with the history of the values of the input variables and the temperature T-06 as the output variable, and training it by adjusting the weights of the connections of the neurons until they gradually converge to the values that cause each input variable value to produce the corresponding theoretical or desired value of temperature T-06;

[0032] - presenting the ANN2 with the history of the values of the input variables and the temperature T-01 as the output variable, and train them by adjusting the weights of the connections of the neurons until they gradually converge to the values that cause each input variable value to produce the corresponding theoretical or desired value of temperature T-01 ;

[0033] - executing ANNs, trained by a learning algorithm, for the estimation of the values of T-01 and T-06 based on the values of the input variables of at least one operational step of the running process, wherein the ANN2 is activated to operate in tandem with ANN1 when T-01 exceeds the threshold value (Tu); and

[0034] - readjusting the variables of the at least one operational step of the process until the estimated values of the temperatures T-06 and T-01 meet the theoretical or desired values with an error that is less than the indicated error; and wherein the theoretical value of T-06 is less than 660°C, the theoretical value of T-01 is equal to or less than 315°C and the Tu is 290°C andthe indicated error is equal or less than 2% for T-01 and T-06.

[0035] The pyrolysis processes for urban and industrial waste are characterised by their non-linear, multivariable, and non-stationary nature, mainly due to the heterogeneity of the treated waste. This characteristic results in a process gas with a variable composition, which makes it difficult to use it to maintain a constant and adequate temperature in the combustion chamber(s) responsible for supplying thermal energy to the pyrolysis reactor.

[0036] The present invention overcomes the limitations of the prior art by efficiently addressing this problem via an innovative control system. This system uses two artificial neural networks (ANNs) acting in tandem to control and optimise the process, achieving significant energy savings compared to conventional control systems.

[0037] BRIEF DESCRIPTION OF DRAWINGS

[0038] The foregoing and other advantages and features will be better understood based on the following detailed description of several embodiments in reference to the attached drawings, which must be interpreted in an illustrative and non-limiting manner and in which:

[0039] Figure 1 shows a schematic diagram of the operation of an artificial neural network. In this case, the network consists of four layers of neurons: the input layer, two hidden layers with a specific number of neurons (01nand 02n) and the output layer, which contains a number of output neurons (Sn) corresponding to the output variables (Vs). The input layer consists of a set of input neurons (En), the number of which corresponds to the number of input variables (Ven). These variables are weighted by the weights of the connections between neurons and a network output signal (Vs) is generated via a transfer function.

[0040] Figure 2 shows a schematic diagram of the pyrolysis plant of the invention.

[0041] Figure 3 shows a schematic diagram of the control process of the pyrolysis plant of the invention.

[0042] List of references

[0043] H-001 - Feed auger

[0044] H-002 - Feed hopper assemblyH-003 - Conveyor belts for the input waste

[0045] H-101 - Combustion chamber

[0046] R-001 - Pyrolysis reactor

[0047] K-001 - Rectification tower 1

[0048] K-002 - Rectification tower 2

[0049] W-001 - Primary condenser

[0050] W-002 - Secondary condenser

[0051] W-003 - Tertiary condenser

[0052] K-003 - Heavy fraction separator

[0053] K-004 - Condensate trap

[0054] D-001 - Heavy fraction decanter

[0055] D-002 - Light fraction decanter

[0056] P-002 - External backflow pump

[0057] D-005 - Safety vent treatment

[0058] D-006 - Gas cleaning

[0059] D-003 - Vacuum system 1

[0060] D-004 - Vacuum system 2

[0061] P-005 - Recirculation pump

[0062] VE-01 - Venturi

[0063] K-008 - Smoke evacuation chimney

[0064] H-102 - Postcombustion of vent and combustion gases

[0065] PREFERRED EMBODIMENT OF THE INVENTION

[0066] The first aspect of the invention relates to a pyrolysis plant for the pyrolysis of industrial and urban plastic waste, comprising at least the following apparatus:

[0067] - means for receiving and conditioning the waste,

[0068] - at least one pyrolysis reactor configured to receive waste and to produce pyrolysis products, these products being condensable hydrocarbon vapours or pyrolysis oil vapours, non-condensable hydrocarbons or process gas and char or coal),

[0069] - at least one combustion chamber configured to provide thermal energy to the pyrolysisreactor(s),

[0070] - at least one rectification tower configured to receive the condensable hydrocarbon vapours or pyrolysis oil vapour and non-condensable hydrocarbons or process gas resulting from the at least one pyrolysis reactor,

[0071] - at least one vapour condensation apparatus configured to receive the condensable hydrocarbon vapours or pyrolysis oil vapour and non-condensable hydrocarbons or process gas from the at least one rectification tower and to produce a flow of light and heavy oil fractions,

[0072] - at least one non-condensable gas scrubbing tower configured to receive non- condensable gases from the at least one vapour condensation apparatus,

[0073] - at least one indirect contact cooling apparatus configured to regulate the temperature of the different apparatus of the plant,

[0074] - at least two tanks configured to receive light and heavy oil fractions resulting from the at least one vapour condensation apparatus and from the at least one rectification tower,

[0075] - at least one tank for receiving wastewater resulting from the at least one vapour condensation apparatus and / or the at least one non-condensable gas scrubbing tower, and

[0076] - at least one postcombustor to combust combustion gases resulting from the at least one combustion chamber,

[0077] wherein the plant is provided with a control system characterised in that it comprises:

[0078] - at least one installation with sensors and measuring instruments associated with the apparatus of the plant whereby they detect the values of the input and output variables of said apparatus;

[0079] - at least one installation for regulating or controlling the operation of the apparatus of the plant;

[0080] - a first analyser for analysing the temperature of the combustion chamber(s) (T-06) in the form of a first artificial neural network (ANN1), such that it emits, from the measured values, control quantities to the regulation and control installations of the apparatus of the plant to adjust T-06 to a set temperature, preferably the set temperature is a temperature lower than 660°C;

[0081] - a second analyser for analysing the temperature at the head of the rectification tower(s) (T-01) in the form of a second artificial neural network (ANN2), such that it emits, from themeasured values, control quantities to the regulation and control installations of the apparatus of the plant to adjust T-01 to a temperature lower than or equal to 315°C;

[0082] wherein the second analyser is activated to operate in tandem with the first analyser when T-01 exceeds a threshold value (Tu), wherein the threshold value is 290°C, and wherein the first and second analyser operating in tandem determine T-06 set temperature.

[0083] The set temperature of the combustion chamber(s) (T-06) calculated by the first and second analyser working in tandem is optimized to a maximum value but without exceeding the 660°C.

[0084] The term “industrial and urban plastic waste” refers to discarded polymer materials from different sources. Industrial plastic waste originates during manufacturing and handling processes in industrial environments, including offcuts, defective products, and industrial packaging. In contrast, urban plastic waste derives from everyday consumption in residential, commercial, and municipal settings, and includes packaging, bags, bottles, and other singleuse products.

[0085] In the pyrolysis plant of the invention the at least one pyrolysis reactor operates as a reboiler or vapor generator, and the at least one rectification column is located immediately downstream of it. In the at least one rectification column, the primary vapours, comprising a mixture of non-condensable Ci-C4components, condensable C5-C32components, and very heavy condensable fractions C32+, rise through the interior of the column. During their ascent, the heavier fractions partially condense within the column and return by gravity to the reactor, thereby contributing to natural reflux and thermal rectification of the overall stream, while the light and intermediate fractions (Ci-C32) continue to rise to the top of the column, where they are withdrawn as the sole outlet stream with the non-condensable fraction (Ci-C4) which is directed to the at least one gas cleaning apparatus after passing through all condensation stages.

[0086] The pyrolysis reactor(s) in the plant of the invention are designed to process industrial and urban plastic waste, operating within a typical temperature range commonly used in the field, generally between 350 and 700 °C. This range allows for the effective processing of various plastic polymers, including polyethylene (PE), polypropylene (PP), polystyrene (PS), and others, depending on the desired end products. In the present plant the reactor(s) operate at a temperature between 370 and 500 °C, preferably between 420 and 460 °C.The temperature T-01 of the at least one rectification tower refers to the temperature at the upper section of the rectification tower, namely the temperature at the headspace of the rectification tower, i.e. head temperature. The head temperature is directly correlated with the vapor composition and, therefore, controls the distillation cut and quality of the condensed pyrolysis oil. Monitoring and adjusting the head temperature thus ensures stable operation and reproducible product characteristics.

[0087] The at least one installation with sensors and measuring instruments has the capability to monitor and record at least the following parameters in view of:

[0088] - Process gas flow rate,

[0089] External gas flow rate,

[0090] - Air flow rate,

[0091] - % O2 in the combustion gases,

[0092] % moisture in the feed materials,

[0093] % of mineral fillers in feed plastics,

[0094] - Weight of fed material,

[0095] - Total volume of generated pyrolysis oils,

[0096] - Total volume of generated wastewater,

[0097] - Total volume of generated process gas,

[0098] - Current weight of the reactor load,

[0099] - Average flow rate of the material fed in the last hour,

[0100] - Average flow rate of N2 fed in the last 5 minutes.

[0101] - Temperature at the head of the rectification tower(s) (T-01),

[0102] - Combustion chamber temperature (T-06),

[0103] - % of plastics in the feed material,

[0104] - Internal temperature of the reactor,

[0105] External backflow rate, and

[0106] - Temperature of the light condensate tank,

[0107] The external gas flow rate refers to LPG or Natural Gas, depending on availability. If the location is connected to a natural gas network, natural gas is used; otherwise, an LPG tank is installed. The combustion chamber burners are dual and can operate with both process gas and external gas. The pyrolysis process is energy self-sustaining, meaning it generates enough process gas to avoid the need for an external source (LPG or Natural Gas). However, during startup, external gas is necessary because, until the pyrolysis regime is reached, either no gas is produced or the gas is very poor in quality. Additionally, when a production batch isnearing completion, a lower-quality gas is produced, requiring the supply of external gas (LPG or Natural Gas).

[0108] The sensors used to monitor the parameters are those commonly known and readily available to a person skilled in the art. Examples include mass flow meters, thermocouples, and pressure transducers, all of which are standard tools in industrial applications for process monitoring and control. Furthermore, the parameters monitored can include the typical and well-established parameters for installations of this type, which are crucial for ensuring optimal operation and efficiency. These parameters are selected based on industry standards and the specific requirements of the pyrolysis process. They typically encompass a wide range of variables, including not only the gas flow rates, temperature, and moisture content as previously mentioned, but also additional factors such as pressure, reaction kinetics, and material composition.

[0109] In the preferred embodiment of the pyrolysis plant, the pyrolysis plant comprises a PLC system that controls the combustion chamber temperature (T-06) during the start-up of the pyrolysis process. This system only controls the start-up of the process and is replaced by the tandem system of the ANNs.

[0110] In the preferred embodiment of the pyrolysis plant, the plant further comprises an external backflow system configured to reintroduce a flow of condensed light pyrolysis oils into the at least one rectification column. The condensed light pyrolysis oil flow is introduced in the higher part of the rectification column but not at the top in such a manner that the temperature at the head of the column, that is T-01, is quickly controlled to acceptable values. This backflow system is activated by the control system when the set temperature is exceeded in the rectification columns (T-01), allowing the cooling of the column head (T-01) without affecting T-06 which allows precise control of T-01 , and the use of the light fraction stream to clean the upward current in the rectification tower, enriching the output with lighter components and preventing the escape of heavy fractions, which results in more efficient usage of energetic resources and yields a higher-purity light oil fraction.

[0111] In the preferred embodiment of the pyrolysis plant, the combustion chamber(s) comprise(s) two independent temperature sensors therein. Said sensors are used to obtain a more accurate temperature of the combustion chamber (T-06).

[0112] In the preferred embodiment of the pyrolysis plant, the means for receiving and conditioningthe waste comprise at least one sealed hopper where the heat from the combustion gases of the combustion chamber is used for the pre-drying of the waste. The use of at least one sealed hopper allows the combustion gases to be used for pre-drying the waste as it is fed into the pyrolysis reactor, resulting in significant energy savings.

[0113] In the preferred embodiment of the pyrolysis plant, the plant comprises at least one integrated rectification column with an internal backflow and an external backflow.

[0114] In the preferred embodiment of the pyrolysis plant, at least one vapour condensation apparatus comprises a plurality of condensers including a pH sensor configured for chlorine detection. The vapour condensation apparatus, which includes a pH sensor configured to detect chlorine, makes it possible to identify the presence of this element in the products resulting from the pyrolysis. Since chlorine is an unwanted component, its early detection activates an alert that facilitates modification of the conditions of the process, either by adjusting the input waste or by correcting possible deficiencies in one of the units.

[0115] In the second aspect of the invention, it relates to a pyrolysis process for the pyrolysis of industrial and urban plastic waste by means of a pyrolysis plant according to the first aspect of the invention, which comprises the operational steps of waste treatment, pyrolysis of the waste and condensation of the outlet vapours of the reactor, characterised in that the temperature of the combustion chamber(s) (T-06) and the temperature of the head of the rectification tower(s) (T-01) is controlled by artificial neural networks (ANNs), and said process comprises:

[0116] - acquiring values of the input variables of at least one operational step of the process by means of sensors and measuring instruments;

[0117] - recording values of the temperature T-06 and the temperature T-01;

[0118] - presenting the ANN1 with the history of the values of the input variables and the temperature T-06 as the output variable, and train them by adjusting the weights of the connections of the neurons until they gradually converge to the values that cause each input variable value to produce the corresponding theoretical or desired value of temperature T-06;

[0119] - presenting the ANN2 with the history of the values of the input variables and the temperature T-01 as the output variable, and train them by adjusting the weights of the connections of the neurons until they gradually converge to the values that cause each input variable value toproduce the corresponding theoretical or desired value of temperature T-01 ;

[0120] - executing the ANNs, trained by a learning algorithm, for the estimation of the values of T-01 and T-06 based on the values of the input variables of at least one operational step of the running process, wherein the ANN2 is activated to operate in tandem with ANN1 when T-01 exceeds the threshold value (Tu); and

[0121] - readjusting the variables of at least one operational step of the process until the estimated values of the temperatures T-06 and T-01 meet the theoretical or desired values with an error that is less than the indicated error; wherein the theoretical value of T-06 is less than 660°C, the theoretical value of T-01 is equal to or less than 315°C, the threshold value (Tu) is 290°C and the indicated error is equal or less than 2% for T-01 and T-06, preferably the error is equal or less than 1.5% for T-01 and equal or less than 0.75% for T-06.

[0122] In the preferred embodiment of the pyrolysis process, the input variables of the ANN1 comprise:

[0123] Process gas flow rate,

[0124] External gas flow rate,

[0125] Air flow rate,

[0126] % O2 in the combustion gases,

[0127] % moisture in the feed materials,

[0128] % of mineral fillers in feed plastics,

[0129] Weight of fed material,

[0130] Total volume of generated pyrolysis oils,

[0131] Total volume of generated wastewater,

[0132] Total volume of generated process gas,

[0133] Current weight of the reactor load,

[0134] Average flow rate of the material fed in the last hour,

[0135] Internal temperature of the reactor, and

[0136] Average flow rate of N2 fed in the last 5 min.

[0137] In the preferred embodiment of the pyrolysis process, the input variables of the ANN2 comprise:

[0138] Combustion chamber temperature (T-06)% moisture in the feed materials,

[0139] % of plastics in the feed material,

[0140] % of mineral fillers in feed plastics,

[0141] Weight of fed material,

[0142] Total volume of generated pyrolysis oils,

[0143] Total volume of generated wastewater,

[0144] Total volume of generated process gas,

[0145] Current weight of the reactor load,

[0146] Average flow rate of the material fed in the last hour,

[0147] Internal temperature of the reactor,

[0148] External backflow rate,

[0149] Temperature of the light condensate tank,

[0150] External gas flow rate, and

[0151] Process gas flow rate,

[0152] The use of these input variables in both ANN1 and ANN2 allows for optimal operation of the pyrolysis process without the need to include quality parameters of the resulting products in the control model. This facilitates the measurement and control of the process in real time, using parameters that are relatively simple to obtain in a method with these characteristics.

[0153] In the preferred embodiment of the pyrolysis process, the ANNs are back-propagation ANNs with at least 4 layers comprising 1 input layer, 2 hidden layers and 1 output layer. In the preferred embodiment of the pyrolysis process, the hidden layers comprise a maximum of 30 perceptrons or neurons. In the preferred embodiment of the pyrolysis process, the learning algorithm used in the ANNs is the Levenberg-Marquardt algorithm.

Claims

CLAIMS1. A pyrolysis plant for the pyrolysis of industrial and urban plastic waste, comprising at least the following apparatus:means for receiving and conditioning the waste,at least one pyrolysis reactor configured to receive waste and to produce pyrolysis products, these products being condensable hydrocarbon vapours or pyrolysis oil vapours, non-condensable hydrocarbons or process gas and char or coal), at least one combustion chamber configured to provide thermal energy to the pyrolysis reactor(s),at least one rectification tower configured to receive the condensable hydrocarbon vapours or pyrolysis oil vapour and non-condensable hydrocarbons or process gas resulting from the at least one pyrolysis reactor,at least one vapour condensation apparatus configured to receive the condensable hydrocarbon vapours or pyrolysis oil vapour and non-condensable hydrocarbons or process gas from the at least one rectification tower and to produce a flow of light and heavy oil fractions,at least one non-condensable gas scrubbing tower configured to receive non- condensable gases from the at least one vapour condensation apparatus,at least one indirect contact cooling apparatus configured to regulate the temperature of the different apparatus of the plant,at least two tanks configured to receive light or heavy oil fractions resulting from the at least one vapour condensation apparatus and from the at least one rectification tower,at least one tank for receiving wastewater resulting from the at least one vapour condensation apparatus and / or the at least one non-condensable gas scrubbing tower, andat least one postcombustor to combust combustion gases resulting from the at least one combustion chamber,wherein the plant is provided with a control system characterised in that it comprises:- at least one installation with sensors and measuring instruments associated with the apparatus of the plant whereby they detect the values of the input and output variables of said apparatus;- at least one installation for regulating or controlling the operation of the apparatus ofthe plant;- a first analyser for analysing the temperature of the combustion chamber(s) (T-06) in the form of a first artificial neural network (ANN1), such that it emits, from the measured values, control quantities to the regulation and control installations of the apparatus of the plant to adjust T-06 to a set temperature, preferably to a temperature lower than 660°C;- a second analyser for analysing the temperature at the head of the rectification tower(s) (T-01) in the form of a second artificial neural network (ANN2), such that it emits, from the measured values, control quantities to the regulation and control installations of the apparatus of the plant to adjust T-01 to a temperature lower than or equal to 315°C;wherein the second analyser is activated to operate in tandem with the first analyser when T-01 exceeds a threshold value (Tu), wherein the threshold value (Tu) is 290°C, and wherein the first and second analyser operating in tandem determine T-06 set temperature.

2. The pyrolysis plant according to claim 1 , characterised in that it further comprises a backflow system configured to reintroduce a flow of condensed light pyrolysis oils into the at least one rectification tower.

3. The pyrolysis plant according to claim 1 , characterised in that the combustion chamber(s) comprise(s) two independent temperature sensors therein.

4. The pyrolysis plant according to claim 1 , characterised in that the means for receiving and conditioning the waste comprise at least one sealed hopper where the heat from the combustion gases of the combustion chamber is used for the pre-drying of the waste.

5. The pyrolysis plant according to claim 1 , characterised in that the plant comprises at least one integrated rectification column with an internal backflow and an external backflow.

6. The pyrolysis plant according to claim 1 , characterised in that at least one gas condensation apparatus comprises a plurality of condensers including a pH sensor configured for chlorine detection.

7. A pyrolysis process for the pyrolysis of industrial and urban plastic waste by means of a pyrolysis plant according to any of the preceding claims, comprising the operational steps of waste treatment, pyrolysis of the waste and condensation of the outlet vapours of the reactor,characterised in that the temperature of the combustion chamber(s) (T-06) and the temperature at the head of the rectification tower(s) (T-01) are controlled by artificial neural networks (ANNs), and said process comprises:- acquiring values of the input variables of at least one operational step of the process by means of sensors and measuring instruments;- recording values of the temperature T-06 and the temperature T-01;- presenting the ANN1 with the history of the values of the input variables and the temperature T-06 as the output variable, and training it by adjusting the weights of the connections of the neurons until they gradually converge to the values that cause each input variable value to produce the corresponding theoretical or desired value of temperature T-06;- presenting the ANN2 with the history of the values of the input variables and the temperature T-01 as the output variable, and train them by adjusting the weights of the connections of the neurons until they gradually converge to the values that cause each input variable value to produce the corresponding theoretical or desired value of temperature T-01 ;- executing the ANNs, trained by a learning algorithm, for the estimation of the values of T-01 and T-06 based on the values of the input variables of at least one operational step of the running process, wherein the ANN2 is activated to operate in tandem with ANN1 when T-01 exceeds the threshold value (Tu);- readjusting the variables of the at least one operational step of the process until the estimated values of the temperatures T-06 and T-01 meet the theoretical or desired values with an error that is less than the indicated error; and wherein the theoretical value of T-06 is less than 660°C, the theoretical value of T-01 is equal to or less than 315°C and the Tu is 290°C and the indicated error is equal or less than 2% for T-01 and T-06, preferably the error is equal or less than 1.5% for T-01 and equal or less than 0.75% for T-06.

8. The process according to the previous claim, characterised in that the input variables of the ANN1 comprise:Process gas flow rate,External gas flow rate,- Air flow rate,% O2 in the combustion gases,% moisture in the feed materials,% of mineral fillers in feed plastics,- Weight of fed material,- Total volume of generated pyrolysis oils,- Total volume of generated wastewater,- Total volume of generated process gas,- Average flow rate of the material fed in the last hour,- Internal temperature of the reactor (T-09), and- Average flow rate of N2 fed in the last 5 min.

9. The process according to any of claims 7 to 8, characterised in that the input variables of the ANN2 comprise:- Combustion chamber temperature (T-06)% moisture in the feed materials,% of plastics in the feed material,% of mineral fillers in feed plastics,- Weight of fed material,- Total volume of generated pyrolysis oils,- Total volume of generated wastewater,- Total volume of generated process gas,- Current weight of the reactor load,- Average flow rate of the material fed in the last hour,- Internal temperature of the reactor,External backflow rate,- Temperature of the light condensate tank,External gas flow rate, and- Process gas flow rate.

10. The process according to any of claims 8 to 10, characterised in that the ANNs are of the back-propagation type with at least 4 layers comprising 1 input layer, 2 hidden layers and 1 output layer.

11. The process according to the preceding claim, characterised in that the hidden layerscomprise a maximum of 30 perceptrons or neurons.

12. The process according to any of claims 8 to 12, characterised in that the learning algorithm is the Levenberg-Marquardt algorithm.

Images