Processing plant, monitoring system and method of monitoring a processing plant

Abstract

The present disclosure relates to a processing plant (1000) comprising: at least one processing circuitry (40), wherein the at least one processing circuitry (40) is configured to execute a plurality of functions, wherein said plurality of functions comprises: a data acquisition function configured to receive sensor data comprising monitoring sensor data acquired by at least one monitoring arrangement (20) of the processing plant and sorting sensor data acquired by at least one sorting arrangement (30) of the processing plant, and process the sensor data to provide unified sensor data; a classification function configured to analyze the unified sensor data by at least one classifier based on at least one classification criterion, and generate at least one classification output for the unified sensor data, and a control function configured to provide output data based on the at least one classification output, wherein the output data comprises at least one of: control data for controlling the processing plant; status data indicative of a status of the processing plant, and property data indicative of at least one property of transported objects. A monitoring system and a method of monitoring the processing plant are also disclosed.

Description

[0001] PROCESSING PLANT, MONITORING SYSTEM AND METHOD OF MONITORING A PROCESSING PLANT

[0002] Technical Field

[0003] The present disclosure relates to a processing plant, a monitoring system for the processing plant, and a method of monitoring the processing plant.

[0004] Background

[0005] In processing plants for material sorting, recycling, and / or manufacturing, there is a growing need for effective surveillance of the overall plant environment. This has led to an increased focus on plant-level monitoring of the processing plant as a whole. Plant-level monitoring typically relies on dedicated monitoring arrangements, such as surveillance cameras or other stationary observation systems.

[0006] Due to the physical scale and structural complexity of modem processing plants, these monitoring arrangements of today often fail to provide a view of or information from all relevant areas. Dedicated monitoring arrangements, while effective within their immediate scope, do not adequately cover blind spots. As a result, critical events such as blockages, equipment malfunctions, or irregularities in material flow may go undetected, which can impair operational reliability, delay detection of faults, and complicate maintenance and safety oversight. As processing plants become increasingly complex and interdependent, these challenges are further amplified.

[0007] Accordingly, there is a need for solutions that alleviate one or more of these technical concerns.

[0008] It is an object of the present disclosure to provide an improved solution that alleviates at least some drawbacks with present solutions or improves over present solutions in at least one aspect or provides an alternative over present solutions in at least one aspect. This and / or at least one other object, which will be implicitly or explicitly detailed in the following disclosure, is solved by one or more of the inventions defined in the claims. Additional advantageous embodiments are described in the following.

[0009] The present disclosure is based on the realization that monitoring of a processing plant can be advantageously enhanced by leveraging or making use of sensor data acquired from any sensor-equipped arrangement within the processing plant, irrespective of its primary operational purpose. Such sensor-equipped arrangements may include, for example, sorting arrangements configured to detect and classify objects based on sorting sensor data; monitoring arrangements, such as: surveillance cameras, object flow sensors, or arrangements comprising any sensor, e.g., thermal, proximity, or other environmental sensors. Although these arrangements are typically designed for localized or task-specific functions, the sensor data they acquire can yield valuable insights into the broader operational state of the processing plant. To enhance monitoring of a processing plant, the present disclosure relates to a process of unifying sensor data, i.e. , a unification process, in which sensor data from the various sensor-equipped arrangements is processed into unified sensor data suitable for monitoring of the processing plant, such as plant-level monitoring. This unification process may for instance address differences in hardware architecture, sensor modality, data format, spatial reference frames, temporal resolution, and semantic labeling of the different sensor-equipped arrangements. By resolving these general incompatibilities, sensor data from different sensor-equipped arrangements can be meaningfully unified, allowing for a more comprehensive and responsive view of plant operations that extends beyond the capabilities of any individual monitoring system.

[0010] Plant-level monitoring may refer to a comprehensive supervision of an entire processing plant, comprising but not limited to, monitoring of throughput, system status, plant control, and / or environmental conditions. It may allow for detailed monitoring of any aspect of the processing plant.

[0011] Sensor-equipped arrangement may refer to any arrangement within the processing plant that comprises at least one sensor arrangement configured to acquire sensor data during operation. Examples of sensor-equipped arrangements comprise sorting arrangements, monitoring arrangements (such as surveillance cameras or surveillance image sensors), transport arrangements, safety systems, and processspecific measurement units.

[0012] Sorting arrangement may refer to a sensor-equipped arrangement having a primary function to sort objects into at least two fractions and / or into at least two different paths. The sorting arrangement may be configured to: acquire sensor data of objects when present in a predetermined detection zone, typically when transported along a path; classify the objects in terms of object properties, such as shape, color, composition, or position, based on the acquired sensor data; and sort the objects in response to a sorting instruction based on at least one object property of the objects.

[0013] Monitoring arrangement may refer to a sensor-equipped arrangement having a primary function to monitor objects when present in a predetermined detection zone within the processing plant. This may include surveillance cameras and / or vision systems and / or other sensor systems comprising a sensor arrangement configured to acquire sensor data of objects when present in a predetermined detection zone, typically transported along a path, to determine e.g. the presence of any objects. In some embodiments, the monitoring arrangement may also be configured to classify the objects in terms of object properties such as shape, color, composition, or position based on the acquired sensor data.

[0014] According to a first aspect, a processing plant is provided. The processing plant comprises at least one sorting system. Each sorting system comprises a transport arrangement configured to transport objects as at least one stream of objects through a plurality of detection zones, the plurality of detection zones comprising a first detection zone and at least one subsequent detection zone arranged downstream of the first detection zone. Each sorting system comprises at least one monitoring arrangement, each monitoring arrangement configured to acquire monitoring sensor data of objects when present in a predetermined one of the plurality of detection zones, and to classify objects with monitoring object classification data. Each sorting system comprises: a plurality of sorting arrangements, wherein the plurality of sorting arrangements comprises Nssorting arrangements, wherein Ns> 10, wherein the plurality of sorting arrangements comprises at least one upstream sorting arrangement and at least one set of Npparallel sorting arrangements configured to directly or indirectly receive respective fractions from the same upstream sorting arrangement, wherein Np> 4, wherein each sorting arrangement comprises an object displacement device. Each sorting arrangement is configured to: acquire sorting sensor data of objects when present in a predetermined one of the plurality of detection zones; provide sorting object classification data indicative of at least one object property of the objects based on the sorting sensor data; provide a sorting instruction indicative of at least one of at least two fractions based on said sorting object classification data, and sort the objects by means of the object displacement device into at least two fractions based on the sorting instruction. Further, of the at least one monitoring arrangement and the plurality of sorting arrangements, only the plurality of sorting arrangements are configured to provide sorting instructions indicative of the at least one fraction to the object displacement devices. The processing plant further comprises: at least one processing circuitry, wherein the at least one processing circuitry is configured to execute a plurality of functions, wherein said plurality of functions comprises: a data acquisition function configured to receive sensor data comprising the monitoring sensor data and the sorting sensor data, and process the sensor data to provide unified sensor data; a classification function configured to analyze the unified sensor data by at least one classifier based on at least one classification criterion, and generate at least one classification output for the unified sensor data, and a control function configured to provide output data based on the at least one classification output, wherein the output data comprises at least one of: control data for controlling the processing plant; status data indicative of a status of the processing plant, and property data indicative of at least one property of transported objects.

[0015] An advantage of unifying sensor data from different sources for monitoring is that it enables comprehensive and deeper analysis of the processing plant. By integrating different types of sensor data, such as image data, temperature readings, and vibration signals from different sources, or sensor data from different sources, into a common reference framework, the processing plant allows for detection of more complex conditions. This may improve accuracy of event classification (e.g., blockages, fire hazards), and / or reduce false positives or false negatives. Unified sensor data also facilitates real-time decision-making, enhanced visualization, and more efficient use of machine learning models that rely on correlated, multi-modal data inputs. Further, an advantage related to the first aspect is that it may provide a monitoring system having an increased coverage of cover blind spots compared to dedicated monitoring systems, such as mechanically enclosed zones, obstructed transport paths, or regions between detection zones of the monitoring arrangements

[0016] As stated above, the data acquisition function is configured to receive sensor data comprising the monitoring sensor data and the sorting sensor data, and process the sensor data to provide unified sensor data. The sensor data may be received by wired and / or wireless connection from any one sensor-equipped arrangement. The sensor data may further comprise operation data of any one device, arrangement, and / or system of the processing plant. For instance, the sensor data may comprise operation data of the at least one transport arrangement. By comprising operation data, the unified sensor data may provide even greater insight into the processing plant, e.g., by allowing monitoring sensor data and / or sorting sensor data to be correlated with the operation data of the at least one device, arrangement, and / or system.

[0017] Further, as stated above, the classification function is configured to analyze the unified sensor data by at least one classifier based on at least one classification criterion, and generate at least one classification output for the unified sensor data. The classification function may utilize one classifier or a plurality of classifiers for analyzing the unified sensor data in terms of one or more different aspects. The analyzing unified sensor data may involve the at least one classifier receiving the unified sensor data as such as input, or any data representation based on the unified sensor data. The classification criterion may comprise one or more predefined parameters, thresholds, or patterns associated with specific features or conditions detectable in the unified sensor data. The classification criterion may comprise measurable physical or chemical properties such as size, shape, color, weight, spectral signature, material composition, or surface texture, as determined from a combination of sensor inputs including, for example, optical cameras, near-infrared (NIR) sensors, X-ray detectors, and weight sensors. The classifier may be configured to apply the classification criterion to distinguish between different categories of objects or materials, or between different states of, or events within, the processing plant. For instance, the classification function may differentiate between plastic, metal, and paper based on material density and spectral reflectance. Classification output may refer to at least one result produced by the at least one classifier based on an analysis of the unified sensor data against the at least one classification criterion. The classification output may indicate or comprise a specific category, class, or status associated with an object, or a state of the processing plant, or an event within the processing plant, a discrete label (e.g., “normal”, “fault”, “anomaly”); a probability distribution over possible classes; and / or a confidence score.

[0018] Furthermore, unifying sensor data from different sources offers significant advantages compared to isolated monitoring of specific aspects. For instance, while specific monitoring might detect a temperature spike or an object blockage separately, unified sensor data allows the processing plant to correlate these events, revealing, for instance, that the blockage caused overheating. In other words, the processing plant may enable root cause analysis. Also, single-aspect monitoring may miss events outside a field of view of the sensor arrangement being used. Unified sensor data enables detection of anomalies that manifest across multiple types of sensor data (e.g., visual distortion + vibration + heat indicating mechanical failure). Unified sensor data may also reduce reliance on any one sensor arrangement, leading to more robust and reliable monitoring, with fewer false positives or missed detections. Further, instead of managing multiple, disjointed monitoring systems, an operator of the processing plant may interact with a single, coherent overview, improving situational awareness and decision-making.

[0019] The output data of the processing plant may provide several key advantages for the operation and optimization of the processing plant. Output data in the form of control data may allow the processing plant to autonomously adjust processing operations in real time, improving responsiveness and reducing the need for manual intervention. In a non-limiting example, the output data may enable a fully automated processing plant, enhancing safety of human personnel. Status data may allow operational transparency, for instance overview of health and / or states of any individual device, arrangement and / or system of the processing plant, or throughput, or fault conditions, which altogether may support predictive maintenance and early fault detection. Property data related to transported objects enhances tracking, reporting, and documentation of object flow and sorting quality, supporting compliance and quality assurance. The output data may also support both automated and operator-led decisions, enhancing process reliability and consistency. Also, the output data may be stored, enabling long-term analysis and optimization of sorting quality, throughput, and resource efficiency, for instance in response to seasonality of objects being processed, any layout changes of the processing plant, or any law or policy changes impacting how processing needs to be performed.

[0020] Seasonality of objects may refer to how a presence of objects within the at least one stream of objects changes over time. For instance, it may change depending on time of the day (night versus day), time of the week, time of the month, or time of the year.

[0021] The processing plant according to the first aspect is detailed further in the following in terms of examples and / or optional features.

[0022] A sorting system may refer to an independent and / or a modular system of the processing plant, comprising: a transport arrangement; at least one monitoring arrangement, and a plurality of sorting arrangements. Each sorting system may include at least one input zone, through which the sorting system receives objects, either in batches or as a continuous stream. Each sorting system may be configured to process objects, either in batches or as a continuous stream. The system may further comprise at least two receiving zones, each configured to receive a respective fraction of the processed objects. The processing plant may comprise one or more such sorting systems. In some embodiments, the processing plant comprises a plurality of sorting systems, each configured to operate in parallel, i.e. , a first sorting system may operate independently of a second sorting system. These parallel operating sorting systems may be distributed across different regions of the plant, enabling parallel object processing along distinct transport paths. Two or more sorting systems may be configured to process objects from different sources, e.g., textile waste, recyclables, and generic waste. However, two or more sorting systems may be configured to process objects from the same source, which may increase redundancy of the processing plant.

[0023] As detailed above, each sorting system comprises a transport arrangement. The transport arrangement is configured to transport objects as at least one stream of objects through a plurality of detection zones, the plurality of detection zones comprising a first detection zone and at least one subsequent detection zone arranged downstream of the first detection zone. A sensor-equipped arrangement may be configured to acquire sensor data of objects when present in a predetermined detection zone of the plurality of detection zones. The plurality of detection zones may comprise two or more detection zones. Two or more detection zones may be spatially separated in a transport direction of objects. Two or more detection zones associated with respective different sensor-equipped arrangements may at least partially overlap or overlap completely. Two or more detection zones may be associated with different stages of an object detection and classification process. For instance, a first detection zone may be associated with an acquisition of preliminary sensor data related to general object characteristics, such as size shape, or velocity, while a second downstream detection zone may be associated with an acquisition of detailed sensor data, such as spectral data, material composition, or surface condition.

[0024] The transport arrangement may comprise at least a first transport arrangement module. The transport arrangement may comprise a plurality of transport arrangement modules. The plurality of transport arrangement modules may be configured to be arranged to provide at least one transport path for objects to be processed and / or sorted. The transport arrangement modules of the transport arrangement may be arranged so as to transport items at least partially in any one of the following configurations: from an input zone to a detection zone; from a first detection zone to a second detection zone; from a detection zone to a sorting zone; from a first sorting zone to a second sorting zone; from a sorting zone to a receiving zone; from a detection zone to a receiving zone; and / or from a receiving zone to a further receiving zone. One or more transport arrangement modules of the transport arrangement may comprise any one of the following: a conveyer belt for transporting the material flow with items to be sorted; a chute arranged for transporting the material flow with items to be sorted; a free-falling segment; an ejection space through which items may travel following displacement or ejection by a force, e.g., supplied by mechanical actuation or by a flow of gaseous media. Any one of the transport arrangement modules may be configured to be controllable by a control signal.

[0025] As detailed above, each sorting system comprises at least one monitoring arrangement. Each monitoring arrangement is configured to acquire monitoring sensor data of objects when present in a predetermined one of the plurality of detection zones, and to classify the objects using monitoring object classification data. Each monitoring arrangement may comprise a sensor arrangement and at least one processing circuitry configured to perform the classification using monitoring object classification data. The sensor arrangement and the processing circuitry of the monitoring arrangement may be communicatively connected via a wired or wireless connection. For example, the sensor arrangement and the processing circuitry of the monitoring arrangement may be integrated into a single device. Alternatively, the sensor arrangement may be a standalone device, with the processing circuitry located externally. The sensor arrangement of the monitoring arrangement may be a camera, i.e. , a sensor configured to capture images in the visible (VIS) spectrum. In some embodiments, the sensor arrangement of the monitoring arrangement is an RGB camera. However, the sensor arrangement of the monitoring arrangement may be of other types of sensor arrangements set forth in this disclosure.

[0026] Sorting object classification data may refer to information derived from sorting sensor data that describes one or more properties of an object, which are relevant for deciding how that object should be sorted. The sorting object classification data may serve as a basis for providing the sorting instruction. The sorting object classification data may include material type, object property, color, geometry, size, weight, surface texture, object identity, time stamp, etc. Monitoring object classification data may refer to information derived from monitoring sensor data that describes one or more properties of an object. The sorting object classification data may include material type, object property, color, geometry, size, weight, surface texture, object identity, time stamp, etc.

[0027] As detailed above, each sorting system comprises a plurality of sorting arrangements. The plurality of sorting arrangements may comprise Nssorting arrangements, wherein Ns> 10, wherein the plurality of sorting arrangements comprises at least one upstream sorting arrangement and at least one set of Npparallel sorting arrangements configured to directly or indirectly receive respective fractions from the same upstream sorting arrangement, wherein Np> 4. As a first example, the plurality of sorting arrangements comprises one upstream sorting arrangement and one set of four parallel sorting arrangements configured to directly or indirectly receive respective fractions from the same upstream sorting arrangement. In other words, although the parallel sorting arrangements may be located at different locations within the processing plant, they may nonetheless be referred to as parallel sorting arrangements, as long as they receive a respective fraction from the same upstream sorting arrangement.

[0028] Directly receive may refer to objects being received, e.g., by means of a transport arrangement, from an upstream sorting arrangement to a downstream sorting arrangement. In other words, the upstream sorting arrangement and the downstream sorting arrangement are arranged in series without any intermediate sorting.

[0029] Indirectly receive may refer to objects being received, e.g., by means of a transport arrangement, from an upstream sorting arrangement to a downstream sorting arrangement via at least one intermediate sorting arrangement. In other words, the upstream sorting arrangement, the at least one intermediate sorting arrangement, and the downstream sorting arrangement are arranged in series.

[0030] The plurality of sorting arrangements comprises Nssorting arrangements, wherein Ns> 10. Here, Nsis an integer parameter. Nsmay be selected in an interval of 10-20, 20-30, 30-40, 40-50, 50-100, or more than 100, or in an interval formed by any combination of these listed intervals.

[0031] The plurality of sorting arrangements comprises at least one upstream sorting arrangement and at least one set of Npparallel sorting arrangements configured to directly or indirectly receive respective fractions from the same upstream sorting arrangement. The at least one set of Npparallel sorting arrangements may comprise two sets of parallel sorting arrangements, or for instance 3, 4, 5, 6, 7, 8, 9, 10, or more sets of parallel sorting arrangements.

[0032] Each set of the at least one set of Npparallel sorting arrangements may comprise Npparallel sorting arrangements. Here, Npis an integer parameter. Npmay be 2, 3, 4, 5, 6, 7, 8, 9, 10, or more or any integer in the interval of 2 to 100 or more. According to one example each set of the at least one set of Npparallel sorting arrangements comprises Npdirectly parallel sorting arrangements, wherein Npis an integer in the interval of 2 to 30 or interval of 2 to 10. Different sets of parallel sorting arrangements may comprise different numbers of sorting arrangements. For instance, a first set of parallel sorting arrangements may comprise 2 sorting arrangements and a second set of parallel sorting arrangements may comprise 4. Different sets of parallel sorting arrangements may be configured to directly or indirectly receive respective fractions from the different upstream sorting arrangements. For instance, a first set of parallel sorting arrangements may configured to directly or indirectly receive respective fractions from the same first upstream sorting arrangement, and a second set of parallel sorting arrangements may be configured to directly or indirectly receive respective fractions from the same second upstream sorting arrangement which is different from the first upstream sorting arrangement. The respective upstream sorting arrangements may be parallel sorting arrangements or arranged in series (directly or indirectly).

[0033] Two parallel sorting arrangements may refer to a first sorting arrangement and a second sorting arrangement, wherein the first sorting arrangement is configured to receive a first fraction from an upstream sorting arrangement and the second sorting arrangement is configured to receive a second fraction from the same upstream sorting arrangement. If the two parallel sorting arrangements directly receive said fractions from the same upstream sorting arrangement without any one fraction being processed and / or sorted by an intermediate sorting arrangement, the two parallel sorting arrangements may be referred to as being directly parallel. If there is at least one intermediate sorting arrangement between the upstream sorting arrangement and one of the two parallel sorting arrangements, the two parallel sorting arrangements may be referred to as being indirectly parallel. Also, as detailed above, each sorting arrangement comprises an object displacement device. The object displacement device may be configured to sort objects present in a sorting zone into at least two fractions. Non-limiting examples of the object displacement device comprises: a robotic arm sorter, a swivel chute, a vacuum pick-and-place sorter, a mechanical or pneumatic pusher, a drop gate, or a gas ejection device fluidly connected to a supply of a gaseous medium, such as air.

[0034] Each sorting arrangement is configured to: acquire sorting sensor data of objects when present in a predetermined one of the plurality of detection zones; provide sorting object classification data indicative of at least one object property of the objects based on the sorting sensor data; provide a sorting instruction indicative of at least one of at least two fractions based on said sorting object classification data, and sort the objects by means of the object displacement device into at least two fractions based on the sorting instruction.

[0035] A sorting instruction may be represented by a signal generated by at least one processing circuitry, such as at least one processing circuitry of a sorting arrangement or the at least one processing circuitry of each sorting system. The sorting instruction may indicate to the object displacement device how to handle a specific object based on sorting object classification data. In other words, the sorting instruction may map an object to a specific fraction (i.e. output category, bin, or stream) based on the sorting object classification data.

[0036] According to one embodiment, at least one of the sorting arrangements of the plurality of sorting arrangements of the at least one sorting system is configured to sort objects into at least two fractions, each fraction corresponding to a respective object category, wherein said object category is based on a first material property set and / or a second material property set. The at least two fractions may be two, three, four, five, six, seven, eight, nine, ten, or more fractions.

[0037] According to one embodiment, the first property set is indicative of at least one of a spectral response of the matter, a material type of the matter, a colour of the matter, a fluorescence of the matter, a ripeness of the matter, a dry matter content of matter, a water content of the matter, a fat content of the matter, an oil content of the matter, a calorific value of the matter, a presence of bones or fishbones of the matter, a presence of pest of the matter, a mineral type of the matter, an ore type of the matter, a defect level of the matter, a detection of hazardous biological materials of the matter, a presence of matter, a non-presence of matter, a detection of multilayer materials of the matter, a detection of fluorescent markers of the matter, a quality grade of the matter, a physical structure of the surface of the matter and molecular structure of the matter.

[0038] According to one embodiment, the second property set is indicative of at least one of a height of the matter, a height profile of the matter, a 3D map of the matter, an intensity profile of reflected and / or scattered light, a volume centre of the matter, an estimated mass centre of the matter, an estimated weight of the matter, an estimated material of the matter, a presence of matter, a non-presence of matter, a detection of isotropic and anisotropic light scattering of the matter, a structure and quality of wood, a surface roughness and texture of the matter and an indication of presence of fluids in the matter.

[0039] According to one embodiment, the at least one monitoring arrangement is configured to classify objects in terms of at least one object category, wherein said object category is based on the first material property set and / or the second material property set.

[0040] According to one embodiment, the control data comprises: data configured to initiate a control operation of the processing plant, which control operation is selected from a group of control operations comprising: control of at least one device and / or system of the processing plant; safety response of the processing plant, such as calling emergency services, such as firefighters; maintenance operation of the processing plant; improving, preferably optimizing, processing of processing plant in at least one aspect, such as throughput, sorting quality, sustainability, energy consumption, carbon dioxide footprint; documentation of objects and a processing thereof, and reconfiguration of processing plant. This may advantageously enable centralized and flexible control of plant operations, improve safety through automated emergency responses, support preventive maintenance and reduce downtime, optimize processing for throughput, quality, and efficiency. Further, it may reduce energy use and carbon footprint. Also, it may advantageously facilitate traceability of object sorting and other aspects through automated documentation. The processing plant may further facilitate easy reconfiguration for changing processing needs.

[0041] According to one embodiment, the status data comprises data relating to at least one of: an overall throughput of the processing plant; the overall throughput of one or more selected compositions such as PVC (Polyvinyl chloride) during a selected time period within the plant (e.g. since a reconfiguration or the start of the plant), throughput within any one detection zone of the plurality of detection zones; a status of at least one device of the processing plant; a layout change of the processing plant; a blockage within any one detection zone of the plurality of detection zones; a fire hazard within any one detection zone of the plurality of detection zones, and a fault of any one device of the processing plant. This may advantageously enable real-time monitoring of plant performance and device health, facilitate early detection of faults, blockages, and fire hazards. This may advantageously further support analysis through zone-specific throughput data, allow for responsive adjustments to layout changes or process conditions. Also, it may enhance operational transparency and decisionmaking efficiency.

[0042] According to one embodiment, the property data comprises data indicative of at least one of: at least one property of the objects being transported and / or sorted, such as the first material property and / or the second material property; at least one statistics of the objects being transported and / or sorted; a sorting quality of at least one sorting arrangement or of at least one sorting system; accumulative sorting information during a predetermined time period; evenness of flow of objects within any one detection zone of the plurality of detection zones. This may advantageously enable detailed tracking of object properties and sorting performance. This may advantageously support statistical analysis for process optimization. This may advantageously support monitoring and improvement of sorting quality over time. Also, it may facilitate evaluation of cumulative performance during specific time periods, and help maintain consistent object flow, reducing blockages and improve processing and / or sorting efficiency.

[0043] According to one embodiment, the processing plant is configured to transmit the output data to a display device and / or a communication device configured to present information based on the output data to an operator of the processing plant. The display device may comprise at least one of a human-machine interface (HMI), a monitor, a control panel, and a touchscreen. The communication device may include wired or wireless communication modules (e.g., Ethernet, Wi-Fi, cellular) for transmitting the data to remote a terminal, such as an operator tablet, a smartphone, or a centralized control system. This configuration may allow an operator of the processing plant to monitor plant conditions, receive alerts, make informed decisions, and adjust operations in real time based on current plant performance and object processing conditions. The display device and / or the communication device may be located within a control room of the processing plant. The display device and / or the communication device may be comprised in a monitoring system. The monitoring system may be controllable from the control room.

[0044] According to one embodiment, said process of the sensor data to provide unified sensor data comprises at least one of: normalizing and / or scaling the values of the received sensor data into new values, which new values are different from the values of the received sensor data; converting the values of the received sensor data into new values according to a frame of reference, wherein the frame of reference for the new values is different from the frame of reference of the values of the received sensor data; providing at least one semantic interpretation to the received sensor data, which at least one semantic interpretation is not present in the received sensor data; categorizing and optionally order the values of the received sensor data into new values according to a predefined system, such as by detection zones or sensor types; normalizing, scaling and / or converting image data preferably with respect to resolution and / or spatial selection; formatting the received sensor data into new sensor data using a structural format different from that of the original sensor data, such as through reformatting, reordering, removal, or nesting of data elements; correlating the received sensor data into new sensor data, which new sensor data is correlated and / or synchronized in time or space; and combining the received sensor data into new sensor data by merging and / or fusing sensor data of e.g. different sensor types.

[0045] Sensor arrangements may produce outputs in different units or ranges (e.g., temperature in Celsius vs. Fahrenheit, or distances in meters vs. millimeters). Normalization brings these diverse values into a consistent scale, such as mapping all values between 0 and 1 , making them easier to compare and analyze together. Scaling can also involve adjusting the magnitude of signals to optimize processing.

[0046] Sensor arrangement may provide sensor data in different spatial or temporal frames of reference. For example, a camera might provide pixel coordinates, while a laser scanner provides distance measurements relative to the device. The step of converting values of received sensor data into new values according to a frame of reference may comprise transforming sensor data so that all measurements align within a single, unified coordinate system or time base. This facilitates integrating data from multiple sources and for accurately interpreting object positions.

[0047] Raw sensor outputs are typically just numerical values without explicit meaning. Semantic interpretation may assign context or labels to data points, such as identifying an object type from patterns in sensor data or tagging detected regions as “obstacle” or “background.” This may add a higher level of understanding to the sensor data, enabling more effective processing and decision-making downstream.

[0048] Sensor data may be organized into categories (e.g., by sensor type or detection zone) and ordered in a specific sequence to facilitate processing. For instance, sensor data from multiple sensor arrangements covering different zones can be grouped accordingly. Alternatively, or in combination, sensor data might be sorted by time or spatial location. This structured approach may improve handling of sensor data.

[0049] Image sensor data often varies in resolution and spatial coverage. By normalizing, scaling and / or converting image data preferably with respect to resolution and / or spatial selection, images may be adjusted to a standard resolution and / or selected regions of interest and / or scale pixel intensity values accordingly. This may allow for consistent image quality and analysis on relevant parts of the scene, improving subsequent image processing tasks.

[0050] Raw sensor data may come in various formats, so it is advantageous to process the sensor data in terms of restructuring, reordering, nesting, or removing elements, into a standardized format suitable for integration, storage, or further analysis.

[0051] Sensor arrangements may operate according to different time references and may sample data at different rates or times. Correlating the received sensor data may comprise aligning sensor data streams so that measurements refer to the same time points or spatial locations. Synchronization may facilitate that data from multiple sensor arrangements are combined meaningfully, preventing errors caused by temporal or spatial mismatches.

[0052] By merging and / or fusing received sensor data from multiple different sensors, a single, cohesive dataset may be acquired. Merging and / or fusing may involve averaging, weighting, or more complex algorithms to exploit complementary sensor strengths, reduce uncertainty, and produce a more accurate and reliable representation of the environment or objects.

[0053] According to one embodiment, the unified sensor data comprises at least one data representation indicative of at least one qualitative aspect selected from the group comprising: being expressed as values relative to a reference frame different from that of the original sensor data; being expressed as normalized or scaled values different from those of the original sensor data; being expressed as image data, optionally with a different resolution and / or spatial selection than that of the original sensor data; having a temporal resolution different from that of the original sensor data; being associated with at least one semantic interpretation of the original sensor data; being categorized according to a predefined categorization system, such as by detection zones or sensor types; being processed using a structural format different from that of the original sensor data, such as through reformatting, reordering, removal, or nesting of data elements; being correlated in time or space; being synchronized in time; being combined by merging the sensor data; being fused by integrating sensor data of different sensor types.

[0054] According to one embodiment, each classifier of the at least one classifier comprises a machine learning model, optionally a neural network or a deep neural network, wherein each classifier is trained on labeled training data to monitor a respective aspect of the at least one aspect of the processing plant. The labeled training data may comprise inputs paired with corresponding known outcomes or classifications. This training enables the classifier to recognize patterns and make predictions or decisions related to a respective aspect of the processing plant. Such aspects may include, for example, monitoring the operational status of equipment, detecting faults or anomalies, classifying types of objects being processed, or assessing sorting quality. The use of neural networks, particularly deep neural networks, allows the classifier to handle complex, high-dimensional data and extract relevant features automatically, thereby improving accuracy and robustness in monitoring the plant’s various operational parameters.

[0055] According to one embodiment, the at least one classifier comprises a first classifier and at least a second classifier, wherein the first classifier is configured to: analyze the unified sensor data based on a first classification criterion, and generate a first classification output for the unified sensor data indicative of a first aspect of the processing plant, and the second classifier is configured to: analyze image data based on a second classification criterion, and generate a second classification output for the unified sensor data indicative of a second aspect of the processing plant. The first aspect and the second aspect may be different, e.g., the first aspect corresponds to blockage detection and the second aspect corresponds to fire-hazard detection. The at least one classifier may comprise any number of classifiers, which may be configured to detect different aspects of the processing plant. It should be understood that the at least one classifier may comprise a plurality of classifiers, such as 2, 3, 4, 5, 10, 50, 100 or more, wherein each classifier is configured to analyze the unified sensor data based on a respective criterion based on an aspect to be monitored in the processing plant.

[0056] According to one embodiment, at least one of the at least one monitoring arrangement and the plurality of sorting arrangements comprises at least one sensor arrangement selected from the group comprising or consisting of: a visible spectrum, VIS, camera; a near-infrared, NIR, sensor; an infrared, IR, sensor; a hyperspectral imaging sensor; a thermal infrared sensor; a 3D stereo vision system; a lidar sensor; a time-of-flight, ToF, sensor; a laser scanner; an X-ray sensor; a polarimetric sensor, laser-induced breakdown spectroscopy, X-ray fluorescence, XRF; a temperature sensor; a humidity sensor; a vibration sensor; a gas sensor; a pressure sensor; a magnetic sensor; an electromagnetic sensor; an electrochemical sensor; a flow sensor; a proximity sensor; an ultrasonic sensor; an acoustic sensor. This may advantageously enable detection of diverse object and environmental properties, which may in turn improve accuracy and precision in classification and sorting. Further, it may enhance safety through early fault and hazard detection.

[0057] According to one embodiment, the at least one monitoring arrangement and the plurality of sorting arrangements each comprises at least one sensor arrangement of the same type. As a non-limiting example, the sensor arrangement of the same type is a camera, for instance an RGB camera. By configuring a processing plant so that any monitoring arrangement and sorting arrangement utilizes at least one sensor arrangement of the same type, it may advantageously facilitate the process of unifying sensor data into unified sensor data.

[0058] According to one embodiment, the data acquisition function is configured to selectively switch, over time, between different subsets of the at least one monitoring arrangement and the plurality of sorting arrangements from where to acquire sensor data from. This may advantageously allow for sequential monitoring via different subsets of the at least one monitoring arrangement and the plurality of sorting arrangements, which may reduce computational burden. The same subset may be revisited regularly, e.g., after a predetermined time interval has elapsed or when prompted, e.g., by an operator of the processing plant. The process of selectively switching, over time, between different subsets of the at least one monitoring arrangement and the plurality of sorting arrangements from where to acquire sensor data from, may be referred to as multiplexed analysis. According to one embodiment, the different subsets of the at least one monitoring arrangement and the plurality of sorting arrangements comprises: a first subset and at least a second subset, wherein each of the at least one monitoring arrangement and the plurality of sorting arrangements of the first subset and the second subset is configured with a sensor arrangement of a first sensor type; and / or wherein each of the at least one monitoring arrangement and the plurality of sorting arrangements of the first subset is configured with a sensor arrangement of a first sensor type and each of the at least one monitoring arrangement and the plurality of sorting arrangements of the second subset is configured with a sensor arrangement of a second sensor type.

[0059] According to one embodiment, at least one sorting arrangement of the plurality of sorting arrangements comprises: at least one processing circuitry of the at least one processing circuitry. By this, a sorting arrangement may perform locally at least one of, or all of, the plurality of functions.

[0060] According to one embodiment, at least one monitoring arrangement of the at least one monitoring arrangement comprises: at least one processing circuitry of the at least one processing circuitry. By this, a monitoring arrangement may perform locally at least one of, or all of, the plurality of functions.

[0061] According to one embodiment, the at least one processing circuitry comprises a remote processing circuitry, such as a cloud server, configured to execute at least one of the plurality of functions, and / or the at least one processing circuitry comprises a local processing circuitry, such as a microcontroller or a computer arranged at a processing facility of the processing plant, configured to execute at least one of the plurality of functions, and / or the at least one processing circuitry comprises a portable processing device, such as a smart device, configured to execute at least one of the plurality of functions. This may advantageously enable flexible and scalable processing by distributing tasks between remote processing circuitry, local processing circuitry, and / or portable processing devices.

[0062] According to one embodiment, the at least one processing circuitry comprises a plurality of processing circuitries configured to be communicatively connected, and wherein a first processing circuitry of the plurality of processing circuitries is configured to execute at least one function of the plurality of functions and at least a second processing circuitry is configured to execute at least one other function of the plurality of functions. This may advantageously allow for the plurality of processing circuitries to be distribute. Execution of the plurality of functions may be performed accordingly.

[0063] According to one embodiment, the at least one processing circuitry is configured to execute at least one of the plurality of functions in real time. This may advantageously enable immediate processing and response to sensor data, improving the speed, accuracy, and efficiency of monitoring, control, and sorting operations. Further, it may support time-critical functions such as fault detection, safety interventions, and dynamic process adjustments.

[0064] According to one embodiment, the unified sensor data comprises image data comprising a two-dimensional array of pixels, each pixel having an associated intensity and / or color value and / or a corresponding geometry transformation mapping the pixel to a location or direction in a predetermined reference system. An intensity value may represent brightness or grayscale level in monochrome images. A color value may be represented in a format such as RGB (Red, Green, Blue) in color images. Each pixel may be associated with a geometry transformation defining a mapping of the pixel to a physical location or direction in a predetermined reference system, such as the coordinate system of the processing plant or a world coordinate system. This is particularly relevant in applications like: stereo vision, where each pixel can be triangulated to a 3D point in space; camera calibration, where lens distortion correction and pose estimation may require a transformation from image coordinates to real-world coordinates; spatial alignment, enabling integration with data from other sensors (e.g. LIDAR, ToF, or robotic systems).

[0065] According to one embodiment, the unified sensor data comprises non-image data, such as a time series of a measured parameter, or metadata defining at least one of: spatial resolution, color space, sensor type, sensor identification information, sensor system location, timestamps, calibration information, and environmental information. This may advantageously enhance context of sensor data by providing essential metadata and time-series measurements. Further, it may enable more precise alignment, synchronization, and calibration between sensor data.

[0066] According to one embodiment, the plurality of functions comprises: a data logging function configured to store the image data, the sensor data, classification output of the at least one classifier, and / or a control action for the processing plant. Specifically, the data logging function may be configured to store this data in at least one of the following: on a local storage medium; on a remote server, such as a cloud- based server, on a portable device. The data logging function may be event-driven, continuous, or scheduled periodically. In other words, the data logging function may be configured to record data for later analysis and / or reference.

[0067] According to one embodiment, the processing plant comprises at least one communication interface, wherein each communication interface is configured to communicatively connect, by wired connection and / or by wireless connection, at least one sorting arrangement of the plurality of sorting arrangements and / or a monitoring arrangement to the at least one processing circuitry. The communication interface may be a wired interface, such as Ethernet, or a wireless interface, such as Wi-Fi.

[0068] According to one embodiment, the processing plant comprises a plurality of sorting systems, where each sorting system is configured to operate independently of other at least one other sorting system of the plurality of sorting systems. A first sorting system may be said to be configured to operate independently of a second sorting system if a shut-down of the second sorting system does not prevent operation of the first sorting system, or vice versa. Further, it may refer to the plurality of sorting systems not exchanging objects mid-processing, i.e. , that an object having been received at an input zone of a first sorting system is not transported to a second sorting system before being sorted into a receiving zone. It should however be understood that a first sorting system may be configured to perform a preliminary sorting and a second sorting system may be configured to perform a subsequent more detailed sorting of objects from the first sorting system.

[0069] According to one embodiment, the predetermined configuration of at least one set of sorting arrangement is configured to operate with at least one of a linear configuration, a branching configuration, and a merging configuration. Linear configuration may refer to two or more sorting arrangements being arranged in series, i.e., a fraction from an upstream sorting arrangement is received by a downstream sorting arrangement. Branching configuration may refer to three or more sorting arrangements, wherein an upstream sorting arrangement provides a first fraction to a first downstream sorting arrangement and a second fraction to a second downstream sorting arrangement. Merging configuration may refer to three or more sorting arrangements, wherein a first upstream sorting arrangement provides a fraction to a downstream sorting arrangement and a second upstream sorting arrangement provides a fraction also to the same downstream sorting arrangement. It should be understood that the complexity of the different configurations is not limited to two or three in number, but may be scaled in complexity depending on the number being implemented in the different configurations. Moreover, the predetermined configuration may comprise any combination of linear configuration, branching configuration, and merging configuration.

[0070] Although the processing plant of the first aspect is disclosed in the context of at least one monitoring arrangement configured to classify objects with monitoring object classification data, the at least one sorting system may differ as previously detailed in that the at least one monitoring arrangement is configured to acquire monitoring data of objects when present in a predetermined one of the plurality of detection zones without performing any classification.

[0071] It should be understood that although the processing plant of the first aspect is disclosed in the context of a plurality of sorting arrangements, the at least one sorting system may differ as previously detailed in that each sorting system of the at least one sorting system comprises only one sorting arrangement, and optionally a plurality of sorting arrangements as detailed previously.

[0072] According to second aspect, a monitoring system for a processing plant is provided. The processing plant comprises: at least one sorting system. Each sorting system comprises: a transport arrangement configured to transport objects as at least one stream of objects through a plurality of detection zones, the plurality of detection zones comprising a first detection zone and at least one subsequent detection zone arranged downstream of the first detection zone. Each sorting system comprises: at least one monitoring arrangement configured to acquire monitoring data of objects when present in a predetermined one of the plurality of detection zones, and classify objects with at least one monitoring object classification. Each sorting system comprises: a plurality of sorting arrangements, comprises Nssorting arrangements, wherein Ns> 10, wherein the plurality of sorting arrangements comprises at least one upstream sorting arrangement and Npparallel sorting arrangements configured to directly or indirectly receive respective fractions from the same upstream sorting arrangement, wherein Np> 4, wherein each sorting arrangement comprises an object displacement device. Each sorting arrangement is configured to: acquire sorting sensor data of objects when present in a predetermined one of the plurality of detection zones; provide sorting object classification data indicative of at least one object property of the objects based on the sorting sensor data; provide a sorting instruction indicative of at least one of at least two fractions based on said sorting object classification data, and sort the objects by means of the object displacement device into at least two fractions based on the sorting instruction. Further, of the at least one monitoring arrangement and the plurality of sorting arrangements, only the plurality of sorting arrangements are configured to provide sorting instructions indicative of the at least one fraction to the object displacement devices. The monitoring system comprises: at least one processing circuitry, wherein the at least one processing circuitry is configured to execute a plurality of functions, wherein said plurality of functions comprises: a data acquisition function configured to receive sensor data comprising the monitoring sensor data and the sorting sensor data, and process the sensor data to provide unified sensor data; a classification function configured to analyze the unified sensor data by at least one classifier based on at least one classification criterion, and generate at least one classification output for the unified sensor data, and a control function configured to provide output data based on the at least one classification output, wherein the output data comprises at least one of: control data for controlling the processing plant; status data indicative of a status of the processing plant, and property data indicative of at least one property of transported objects.

[0073] The monitoring system may be adapted with any feature detailed in the context of processing plant of the first aspect or any embodiment thereof.

[0074] According to one embodiment, the monitoring system is configured to be communicatively connected to each sensor arrangement of the at least one monitoring arrangement and the plurality of sorting arrangements. The monitoring system may comprise each sensor arrangement of the at least one monitoring arrangement and each sensor arrangement of the plurality of sorting arrangements. The monitoring system may comprise at least one communication interface for communicative connection with each sensor arrangement of the at least one monitoring arrangement and each sensor arrangement of the plurality of sorting arrangements.

[0075] The monitoring system may comprise a storage medium for storing data, e.g., any data stored by means of a data logging function detailed in the context of the processing plant of the first aspect of the disclosure or any embodiment thereof.

[0076] The monitoring system may comprise a human-machine interface enabling an operator to interact with the monitoring system. The monitoring system may comprise at least one display configured to display at least a portion of the output data, or a representation based on the output data and illustrating an aspect thereof. The monitoring system may comprise an alarm system configured to alert an operator via a visual indicator and / or an audible indicator based on the output data. The monitoring system may be configured to be communicatively connected to an emergency system, such as a sprinkler system, and activate the same based on the output data. The monitoring system may be configured to automatically call emergency services, such as firefighters, based on the output data.

[0077] According to one embodiment, the monitoring system is configured as a supervisory control and data acquisition, SCADA, system. As a SCADA system, the monitoring system may be configured to acquire real-time sensor data from the at least one monitoring arrangement and the plurality of sorting arrangements. Further, the SCADA system may monitor operational status of devices, arrangements, and systems within the processing plant, and provide centralized visualization and control functionalities. The SCADA system may include communication interfaces for wired and / or wireless connectivity to sensors, actuators, and processing circuitry. The SCADA system may further comprise data logging capabilities to store historical data, alarm management to detect and notify of abnormal conditions such as blockages or fire hazards, and interfaces for operator interaction via control room displays or portable devices. The SCADA system enables efficient supervision, fault detection, safety management, and process optimization of the sorting and processing operations.

[0078] According to a third aspect, a method of controlling a processing plant is provided. The processing plant comprises: at least one sorting system. Each sorting system comprises: a transport arrangement configured to transport objects as at least one stream of objects through a plurality of detection zones, the plurality of detection zones comprising a first detection zone and at least one subsequent detection zone arranged downstream of the first detection zone. Each sorting system comprises: at least one monitoring arrangement configured to acquire monitoring data of objects when present in a predetermined one of the plurality of detection zones, and classify objects with at least one monitoring object classification. Each sorting system comprises: a plurality of sorting arrangements, comprises Nssorting arrangements, wherein Ns> 10, wherein the plurality of sorting arrangements comprises at least one upstream sorting arrangement and Npparallel sorting arrangements configured to directly or indirectly receive respective fractions from the same upstream sorting arrangement, wherein Np> 4, wherein each sorting arrangement comprises an object displacement device. Each sorting arrangement is configured to: acquire sorting sensor data of objects when present in a predetermined one of the plurality of detection zones; provide sorting object classification data indicative of at least one object property of the objects based on the sorting sensor data; provide a sorting instruction indicative of at least one of at least two fractions based on said sorting object classification data, and sort the objects by means of the object displacement device into at least two fractions based on the sorting instruction. Further, of the at least one monitoring arrangement and the plurality of sorting arrangements, only the plurality of sorting arrangements are configured to provide sorting instructions indicative of the at least one fraction to the object displacement devices. The method comprises: receiving sensor data comprising the monitoring sensor data and the sorting sensor data, and processing the sensor data to provide unified sensor data; analyzing the unified sensor data by at least one classifier based on at least one classification criterion, and generating at least one classification output for the unified sensor data, and providing output data based on the at least one classification output, wherein the output data comprises at least one of: control data for controlling the processing plant; status data indicative of a status of the processing plant, and property data indicative of at least one property of transported objects.

[0079] The method may be adapted with any feature detailed in the context of processing plant of the first aspect or any embodiment thereof or in the context of the monitoring system or any embodiment thereof.

[0080] According to a fourth aspect, a computer program is provided. The computer program comprises instructions which, when the program is executed by at least one processing circuitry, cause the at least one processing circuitry to carry out the method according to the third aspect or any embodiments thereof.

[0081] According to a fifth aspect, a computer-readable storage medium is provided. The computer-readable storage medium comprises instructions which, when executed by at least one processing circuitry, cause the at least one processing circuitry to carry out the method according to the third aspect or any embodiments thereof.

[0082] Effects and features of the second and third and fourth and fifth aspects are largely analogous to those described above in connection with the first aspect. Embodiments mentioned in relation to the first aspect are largely compatible with the second and third and fourth and fifth aspects. It is further noted that the inventive concepts relate to all possible combinations of features unless explicitly stated otherwise.

[0083] The invention is defined by the appended independent claims, with embodiments being set forth in the appended dependent claims, in the following description and in the drawings. It is to be understood that this disclosure is not limited to the particular component parts of the device described or steps of the methods described as such device and method may vary. It is also to be understood that the terminology used herein is for purpose of describing particular embodiments only and is not intended to be limiting. It must be noted that, as used in the specification and the appended claims, the articles "a", "an", "the", and "said" are intended to mean that there are one or more of the elements unless the context clearly dictates otherwise. Thus, for example, reference to "a unit" or "the unit" may include several devices, and the like. Furthermore, the words "comprising", "including", "containing" and similar wordings do not exclude other elements or steps.

[0084] Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to “a / an / the [element, device, component, means, step, etc.]” are to be interpreted openly as referring to at least one instance of said element, device, component, means, step, etc., unless explicitly stated otherwise.

[0085] Brief Description of the Drawings

[0086] The invention will in the following be described in more detail with reference to the enclosed drawings, wherein:

[0087] Fig. 1a schematically illustrates a processing plant according to some embodiments;

[0088] Fig. 1 b schematically illustrates a sorting system according to some embodiments;

[0089] Fig. 2a schematically illustrates a sorting system according to some embodiments;

[0090] Fig. 2b schematically illustrates a sorting system according to some embodiments;

[0091] Fig. 3 shows a flow chart of a method of monitoring a processing plant according to some embodiments; Fig. 4 illustrates a monitoring arrangement and a sorting arrangement according to some embodiments, and

[0092] Fig. 5 shows a graph over precision of statistics provided by the method according to some embodiments.

[0093] Description of Embodiments

[0094] Hereinafter, the principle of the present disclosure will be described with reference to illustrative embodiments. It should be understood that all these embodiments are given merely for the person skilled in the art to better understand and further practice the present disclosure, but not for limiting the scope of the present disclosure. For example, features illustrated or described as part of one embodiment may be used with another embodiment to yield still a further embodiment. In the interest of clarity, not all features of an actual implementation are described in this specification.

[0095] The disclosed subject matter will now be described with reference to the attached drawings. Various structures, systems and devices are schematically depicted in the drawings for purposes of explanation only and so as to not obscure the description with details that are well known to those skilled in the art. Nevertheless, the attached drawings are included to describe and explain illustrative examples of the disclosed subject matter.

[0096] Fig. 1a schematically illustrates a processing plant according to some embodiments. The processing plant 1000 comprises at least one sorting system. Each sorting system is configured to sort objects being received as input into at least two fractions. In some embodiments, the processing plant 1000 comprises a plurality of sorting systems 100-1 , 100-i, 100-Np. Generally, each sorting system is configured to operate independently of at least one other sorting system, preferably any sorting system. This allows for increased redundancy, allowing for any one sorting system to be shut down for maintenance while allowing the other sorting systems to operate nominally. The different sorting systems may be configured to process and sort different types of categories. As a non-limiting example, a first sorting system may be configured to process and sort general waste whereas a second sorting system may be configured to process and sort textile waste. Further, a first sorting system may be configured to perform general sorting into at least two fractions f 1 , fn, whereas a second sorting system is configured to receive one of said at least two fractions for more specialized processing and sorting.

[0097] The present disclosure is however not limited in terms of number of sorting systems. Rather, a processing plant according to the present disclosure may comprise N sorting systems, wherein N is equal to or greater than 2, optionally selected in an interval of 2-10 or more. Any one sorting system is configured to sort into n fractions, wherein n is equal to or greater than 2, optionally selected in an interval of 2-100 or more, or in an interval of 20 - 60, or in an interval of 30 - 50.

[0098] Fig. 1 b schematically illustrates a sorting system according to some embodiments. The sorting system 100-i comprises a transport arrangement 10 configured to transport objects as at least one stream of objects through a plurality of detection zones DZ, the plurality of detection zones comprising a first detection zone and at least one subsequent detection zone arranged downstream of the first detection zone. The transport arrangement 10 may in some embodiments comprise at least one transport arrangement module or a plurality thereof. A transport arrangement module may be a conveyor belt. The transport arrangement 10 may thus comprise a plurality of conveyor belts configured to transport objects through the plurality of detection zones DZ. The transport arrangement 10 may comprise other types of transport arrangement modules, such as a chute or a free-falling segment. The transport arrangement 10 may be configured to be controllable so as to allow adjustment of a transport speed of objects being transported.

[0099] In some embodiments, all objects are transported through a first detection zone, and then based on sorting by at least one sorting arrangement, the objects will be diverted into at least two fractions, each of which may move through different sets of detection zones. In other words, the plurality of detection zones may be arranged in a branching configuration. Alternatively, or in combination, some detection zones may be arranged in a linear configuration, i.e. , all objects moving through a first detection zone also move through a subsequent second detection zone. Alternatively, or in combination, some detection zones may be arranged in a merging configuration, i.e., objects moving through a first detection zone and objects moving through a second detection zone are all moved through a third merging detection zone.

[0100] The sorting system 100-i also comprises at least one monitoring arrangement 20. Although only one monitoring arrangement is depicted in Fig. 1 b, the sorting system 100-i may comprise a plurality of monitoring arrangements 20. Each monitoring arrangement comprises a sensor arrangement 21 configured to acquire monitoring sensor data of objects when present in a predetermined one of the plurality of detection zones. The monitoring arrangement may optionally be configured to classify objects with monitoring classification data. The sensor arrangement 21 may comprise an RGB camera and / or any other type of sensor arrangement set forth in this disclosure. Each monitoring arrangement may be configured for a primary function of monitoring objects being transported through a predetermined detection zone DZ.

[0101] The sorting system 100-i also comprises at least one sorting arrangement 30, optionally a plurality of sorting arrangements 30. The plurality of sorting arrangements 30 may comprise Np parallel sets of sorting arrangements, each set of sorting arrangements 30 comprising Ns sorting arrangements 30 arranged according to a predetermined configuration. In some embodiments, Np is equal to or greater than 2, but may be selected from an interval of 2 to 50 or more. In some embodiments, Ns is equal to or greater than 1 , but may be selected from an interval of 1 to 50 or more. The predetermined configuration may be at least one of a linear configuration, a branching configuration and a merging configuration.

[0102] Each sorting arrangement 30 comprises an object displacement device 32. The object displacement device 32 may enable a sorting arrangement 30 to sort objects into at least two fractions. Here, the at least two fractions may be an actively sorted fraction and a passively sorted fraction, or a plurality of actively sorted fractions, optionally in combination with a passively sorted fraction.

[0103] Active sorting may refer to sorting as a result of operation of the object displacement device 32. Passive sorting may refer to sorting as a result of no operation of the object displacement device 32.

[0104] As a non-limiting example, the object displacement device 32 is a robotic arm sorter configured to pick up specific objects and move them into at least one actively sorted fraction. Any objects not picked up by the robotic arm sorter may be transported past the robotic arm sorter, thus resulting in passively sorted fraction. Non-limiting examples of the object displacement device 32 comprises: a robotic arm sorter, a swivel chute, a vacuum pick-and-place sorter, a mechanical or pneumatic pusher, a drop gate, or a gas ejection device fluidly connected to a supply of a gaseous medium, such as air.

[0105] Each sorting arrangement 30 comprises a sensor arrangement 31. The sorting arrangement 30 is configured to acquire sorting sensor data of objects by means of the sensor arrangement 31 when present in a predetermined one of the plurality of detection zones. The sorting arrangement 30 is further configured to provide sorting object classification data indicative of at least one object property of the objects based on the sorting sensor data. The sorting arrangement 30 is further configured to provide a sorting instruction indicative of at least one of at least two fractions based on said sorting object classification data. The sorting arrangement 30 is further configured to sort the objects by means of the object displacement device 32 into at least two fractions based on the sorting instruction.

[0106] As a non-limiting example, the sorting system is configured to sort bottles based on color into a first fraction comprising transparent bottles and a second fraction comprising colored bottles. A sensor arrangement 31 of a sorting arrangement 30 may acquire sorting sensor data of the bottles when moving through a detection zone, e.g., by means of a conveyor belt. The sorting sensor data may indicate which bottles are transparent and which bottles are colored. The sorting instruction may indicate that transparent bottles are to be sorted into the first fraction and colored bottles are to be sorted into the second fraction. The sorting arrangement 30 may provide the sorting instruction to the object displacement device 32, e.g., a gas ejection device, a robotic arm sorter, or the like, causing the object displacement device 32 to sort the bottles accordingly.

[0107] Although the sorting system comprises generally two different types of sensor- equipped arrangements, i.e., a monitoring arrangement and a sorting arrangement, the sorting system may be adapted so that, of the at least one monitoring arrangement 20 and the at least one sorting arrangement 30, only the at least one sorting arrangement 30 is configured to provide sorting instructions indicative of the at least one fraction to the object displacement devices. In other words, the at least one monitoring arrangement 20 is configured to acquire monitoring sensor data for a primary purpose of monitoring objects, whereas the at least one sorting arrangement 30 is configured to acquire sorting sensor data for a primary purpose of sorting objects. As such, the monitoring sensor data and the sorting sensor data are acquired for different primary purposes.

[0108] The processing plant 1000 further comprises at least one processing circuitry, wherein the at least one processing circuitry is configured to execute a plurality of functions. Here, at least one processing circuitry may be only one processing circuitry or a plurality of processing circuitries. The plurality of processing circuitries may be distributed across different devices, arrangements, and / or systems. For instance, the at least one processing circuitry 40 comprises a remote processing circuitry, such as a cloud server, configured to execute at least one of the plurality of functions. Alternatively, or in combination, the at least one processing circuitry 40 comprises a local processing circuitry, such as a microcontroller or a computer arranged at a processing facility of the processing plant 1000, configured to execute at least one of the plurality of functions. Alternatively, or in combination, the at least one processing circuitry 40 comprises a portable processing device, such as a smart device, configured to execute at least one of the plurality of functions. The at least one processing circuitry 40 is communicatively connected to each sensor arrangement 21 , 31 of the at least one monitoring arrangement 20 and the at least one sorting arrangement 30, e.g., by means of a communication interface for wired communication and / or wireless communication. In the case of a plurality of processing circuitries, the communication interface is configured to enable communicative connection between the plurality of processing circuitries, e.g., by wired communication and / or wireless communication. As such, the configuration of the at least one processing circuitry configured to execute a plurality of functions is flexible to accommodate different configurations of the processing plant.

[0109] The plurality of functions comprises: a data acquisition function, a classification function, and a control function. The plurality of functions optionally comprises a data logging function.

[0110] The data acquisition function is configured to receive sensor data comprising the monitoring sensor data and the sorting sensor data, and process the sensor data to provide unified sensor data. In other words, sensor data from each sensor arrangement of any sensor-equipped arrangement, be it a monitoring arrangement or a sorting arrangement, is received and processed to provide unified sensor data. This process of providing unified sensor data, also referred to as unification process, may for instance address differences in hardware architecture, sensor modality, data format, spatial reference frames, temporal resolution, and semantic labeling of the different sensor-equipped arrangements. By resolving these general incompatibilities, sensor data from different sensor-equipped arrangements can be meaningfully unified, allowing for a more comprehensive and responsive view of plant operations that extends beyond the capabilities of any individual monitoring system. The classification function is configured to analyze the unified sensor data by at least one classifier based on at least one classification criterion, and generate at least one classification output for the unified sensor data. In some embodiments, the at least one classifier comprises a plurality of classifiers, wherein at least two classifiers are configured to analyze different aspects of objects, states, or conditions of the processing plant, and generate corresponding classification outputs. A classifier of the at least one classifier or the plurality of classifiers may be configured to analyze the unified sensor data, as a whole or selectively depending on source of sensor data, and generate a classification output indicative of at least one of the following non- exhaustive list comprising: blockage, fire hazard, equipment malfunction, throughput, and sorting quality. The classifiers may be configured to analyze the unified sensor data for any other purpose set forth in this disclosure.

[0111] The control function is configured to provide output data based on the at least one classification output, wherein the output data comprises at least one of: control data for controlling the processing plant; status data indicative of a status of the processing plant, and property data indicative of at least one property of transported objects. In other words, the processing plant may advantageously utilize sensor data of any sensor-equipped arrangement of the processing plant to provide unified sensor data, which allows for a more comprehensive monitoring of the processing plant.

[0112] Fig. 2a schematically illustrates a sorting system according to some embodiments. The sorting system comprises a plurality of monitoring arrangements 20a, 20b, 20c. The sorting system comprises a plurality of sorting arrangements 30a, 30b, 30c, 30d, 30e, 30f. A first sorting arrangement 30a is configured to receive objects to be sorted, and sort the objects into two fractions. A first fraction from the first sorting arrangement 30a is transported to a second sorting arrangement 30b. A second fraction from the first sorting arrangement 30a is transported to a third sorting arrangement 30c. The second sorting arrangement 30b is configured to also sort objects into two fractions. A first fraction of the second sorting arrangement 30b is transported to a fourth sorting arrangement 30d. A second fraction of the second sorting arrangement 30b comprises objects which has been erroneously sorted by the first sorting arrangement 30a or objects which otherwise are not suitable to be sorted into the first fraction to the fourth sorting arrangement 30d. The second fraction of the second sorting arrangement 30b is transported to the third sorting arrangement 30c, being transported through a detection zone from where a first monitoring arrangement 20a is configured to monitor. In other words, the monitoring arrangement 20a may monitor the objects of the second fraction from the second sorting arrangement 30b for the purpose of determining characteristics of the objects which were erroneously sorted or not suitably sorted into the first fraction of the second sorting arrangement 30b. The fourth sorting arrangement 30d is configured to sort objects into two fractions, wherein one fraction is transported to a fifth sorting arrangement 30e for sorting objects into further two fractions. The third sorting arrangement 30c receives fractions from the first sorting arrangement 30a and the second sorting arrangement 30b. The third sorting arrangement 30c is configured to sort objects into only one fraction which is transported to a sixth sorting arrangement 30f. The sixth sorting arrangement 30f may be configured to loop objects back to the third sorting arrangement 30c, e.g., for the purpose of preventing a blockage at the sixth sorting arrangement 30f. The sixth sorting arrangement 30f is configured to sort objects into two fractions. A second monitoring arrangement 20b is configured to monitor objects being transported back to the third sorting arrangement 30c. A third sorting arrangement 30c is configured to monitor objects being transported into a first fraction from the sixth sorting arrangement 30f. As a result, the sorting system, by means of the plurality of sorting arrangements 30a- 30f, is configured to sort objects into five different fractions, f1 , f2, f3, f4, f5.

[0113] As detailed above, the plurality of sorting arrangement comprises Nssorting arrangements, wherein Ns> 10, wherein the plurality of sorting arrangements comprises at least one upstream sorting arrangement and at least one set of Npparallel sorting arrangements configured to directly or indirectly receive respective fractions from the same upstream sorting arrangement, wherein Np> 4. In the example embodiment of a sorting system of a processing plant depicted in Fig. 2a, the number of sorting arrangements in the plurality of sorting arrangements is ten, i.e., Ns= 10. Further, the number Npof directly parallel sorting arrangements in at least one set of parallel sorting arrangements is four, i.e., Np= 4, which set includes the sorting arrangements 30c, 30b, 30g, and 30i, each of which are configured to directly receive respective fractions from the same upstream sorting arrangement 30a. Further, Fig. 2a also illustrates sorting arrangements 30d, 30e, each of which are configured to indirectly receive a fraction of the first sorting arrangement 30a. It should however be understood that a sorting arrangement configured to receive a fraction from an upstream sorting arrangement does not necessitate that the sorting arrangement receives all objects from the upstream sorting arrangement, only that it can receive at least some objects from this fraction. Seen as a whole, the sorting system of Fig. 2a is configured to sort objects into nine different fractions, f 1 , f2, f3, f4, f5, f6, f7, f8, f9.

[0114] Fig. 2b schematically illustrates a sorting system according to some embodiments. The sorting system of Fig. 2b differs from the sorting system of Fig. 2a in that the upstream sorting arrangement 30a is configured to sort into only two fractions, which are directly received by a first downstream sorting arrangement 30b and a second downstream sorting arrangement 30c, respectively. Seen as a whole, the sorting system of Fig. 2b is configured to sort objects into at least five different fractions, f1 , f2, f3, f4, f5.

[0115] Fig. 3 shows a flow chart of a method of monitoring a processing plant according to some embodiments. A plurality of different sensor-equipped arrangements, such as different monitoring arrangements 20a, 20b, and different sorting arrangements 30a, 30b, are communicatively connected with the at least one processing circuitry 40. The at least one processing circuitry, which may be one processing circuitry or a plurality of processing circuitries distributed across different devices, arrangements, and systems, may implement the method. The method comprises receiving sensor data comprising the monitoring sensor data and the sorting sensor data. The monitoring sensor data is received from at least one monitoring arrangement or from a plurality of monitoring arrangements 20a, 20b. The sorting sensor data is received from at least one sorting arrangement or from a plurality of sorting arrangements 30a, 30b. The method comprises processing the sensor data to provide unified sensor data. Here, the sensor data may be processed according to a unification process set forth in this disclosure. The method comprises analyzing the unified sensor data by at least one classifier based on at least one classification criterion, and generating at least one classification output for the unified sensor data. In Fig. 3, N classifiers are illustrated, which are configured to generate a corresponding N classification outputs N. Each classifier may however be configured to generate a plurality of different classification outputs. The method comprises providing output data based on the at least one classification output, wherein the output data comprises at least one of: control data for controlling the processing plant; status data indicative of a status of the processing plant, and property data indicative of at least one property of transported objects. The output data may be transmitted to a device, arrangement, or system of the processing plant, such as a monitoring system, or to a remote server external to the processing plant. The processing plant or the remote server may store the output data and / or analyze the output data and / or initiate a control operation in response to the output data.

[0116] The output data may comprise control data. As a non-limiting example, the control data comprises: data configured to initiate a control of at least one device and / or system of the processing plant and / or data configured to initiate a safety response of the processing plant. For instance, one classifier may be configured to detect fire hazards, and the output data may indicate the existence of a fire hazard, for which a safety response such as calling an emergency service (e.g., firefighters) is initiated. Alternatively, or in combination, a classifier may be configured to detect blockages, and the output data may indicate the existence of a blockage at a predetermined location. The control data may control the processing plant accordingly to resolve the blockage. Alternatively, or in combination, a classifier may be configured to detect an equipment malfunction, and the output data may indicate the existence of an equipment malfunction. The control data may control the processing plant accordingly, for instance shutting down impacted portions of the processing plant, to allow for maintenance.

[0117] The output data may comprise status data, the status data comprises data relating to at least one of: an overall throughput of the processing plant 1000; throughput within any one detection zone DZ of the plurality of detection zones DZ; a status of at least one device of the processing plant; a layout change of the processing plant 1000; a blockage within any one detection zone DZ of the plurality of detection zones DZ; a fire hazard within any one detection zone DZ of the plurality of detection zones DZ, and a fault of any one device of the processing plant 1000.

[0118] The output data may comprise property data, the property data comprises data indicative of at least one of: at least one property of the objects being transported and / or sorted; at least one statistics of the objects being transported and / or sorted; a sorting quality of at least one sorting arrangement 30 or of at least one sorting system; accumulative sorting information during a predetermined time period; evenness of flow of objects within any one detection zone DZ of the plurality of detection zones DZ.

[0119] As detailed above, the classification function is configured to analyze the unified sensor data using at least one classifier. The classifier may be a trained classifier. The classifier may implement a machine learning model. The classifier may be configured to process the unified sensor data in order to generate a classification output for the analyzed unified sensor data, which classification output may be a label or a score indicative of a monitored aspect.

[0120] In some embodiments, at least one of the at least one classifier is implemented using a supervised machine learning algorithm, such as a support vector machine (SVM), a random forest, or a deep neural network (e.g., a convolutional neural network or recurrent neural network). The choice of classifier may depend on the type and complexity of the sensor data to be analyzed.

[0121] The unified sensor data comprises sensor data from multiple sensor-equipped arrangements, such as monitoring arrangement or sorting arrangements, which may comprise sensor arrangements including but not limited to, VIS or NIR imaging sensors, spectroscopy systems, accelerometers, temperature sensors, gyroscopes, microphones, or other environmental or physiological sensors, etc. or any other sensor arrangement set forth in this disclosure. This sensor data is processed into the unified sensor data. The unified sensor data may comprise image data, temporal characteristics, or statistical measures. The at least one classifier may be configured to extract features from the unified sensor data.

[0122] The at least one classifier may be trained using a labeled dataset comprising historical sensor data and / or corresponding ground-truth classifications. This training data is typically representative of expected operating conditions or relevant aspects to monitor. The labeled dataset may comprise a sufficient number of examples to allow the at least one classifier to generalize to unseen data. During training, model parameters of the at least one classifier may be adjusted to minimize a loss function that quantifies a classification error.

[0123] During operation, the at least one classifier receives as input the unified sensor data or features derived therefrom. The at least one classifier is configured to analyze the input based on its configuration and generate at least one classification output, which may be: a discrete label (e.g., “normal”, “fault”, “anomaly”); a probability distribution over possible classes; a confidence score. The at least one classification output is comprised in the output data, which may be used to monitor states of the processing plant or properties of objects being processed and / or sorted, or trigger responses, such as alerts, logging events, or control actions.

[0124] Fig. 4 illustrates a perspective schematic view of a monitoring arrangement 20 and a sorting arrangement 700 according to some embodiments. The sorting arrangement 700 is fed with objects 710. The objects 710 is conveyed through a detection zone 720. However, the objects may be provided through the detection zone by any suitable means or manually without any technical means. The sorting arrangement 700 comprises a sensor arrangement and an object displacement device. In some embodiments, the sensor arrangement comprises a light source arrangement 730 and a NIR spectroscopy system 222. The NIR spectroscopy system is adapted to receive and analyze 732, from the light source arrangement 730, which is reflected and / or scattered by the objects in the detection zone 720. Hence, the NIR spectroscopy system 222 typically acquires a spectrum of the objects from the stream of objects that is conveyed through the detection zone 720. The NIR spectroscopy system 222 of the sorting arrangement 700 is configured to discriminate objects from other material and / or objects based on the acquired spectrum. In other words, the NIR system 222 is typically set up such that a specific type of objects is discriminated form other types of the objects owing from its spectrum.

[0125] The sensor arrangement of the sorting arrangement 700 may further comprise a spectroscopy system 760, such as a NIR spectroscopy system, configured to acquire a spectrum of the objects originating from the stream of objects that is conveyed through the detection zone 720. In some embodiments, the object displacement device is an ejection unit configured to eject objects by means of a flow of gaseous media from a source of gaseous media, such as air. The ejection unit 224 of the sorting arrangement 700 may be further configured to divert said objects originating from the stream of objects based on the acquired spectrum, thereby sorting the objects based on color. The spectroscopy system 760 may through the acquired spectrum determine the different colors of the objects that is conveyed through the detection zone 720. An advantage of determining the colors of the objects is that it may be sorted into different fractions.

[0126] The sensor arrangement of the sorting arrangement 700 may further comprise a laser triangulation system 740 configured to determine height information of objects that is conveyed through the detection zone 720. The ejection unit 224 of said sorting arrangement 700 may be further configured to divert said objects based on the determined height information. The laser triangulation system 740 is typically configured to emit a line of laser light 742 towards the detection zone 720. The depicted laser triangulation system 740 includes a camera-based sensor arrangement 744 configured to receive and analyze light 746 which is reflected and / or scattered by the objects in the detection zone 720. By means of the laser triangulation system 740 the arrangement 700 may be able to detect objects that the NIR spectroscopy system 222 have difficulties to detect. By detecting height differences on a conveyor belt used to convey the stream of objects being sorted, the sorting arrangement may combine such height information with the acquired information from the NIR spectroscopy system 222 to determine if there is any object that is hard to detect on the conveyor belt.

[0127] The sensor arrangement of the sorting arrangement 700 may further comprise a camera 750 configured to acquire images of objects originating from the stream of objects that is conveyed through the detection zone 720. The sorting arrangement 700 may comprise a classifier, such as an artificial neural network, in combination with the camera 750. Such artificial neural network may be configured to detect different characteristics of objects that is conveyed through the detection zone 720 based on images acquired by the camera 750 or from any sensor data acquired by any other system of the sensor arrangement. The ejection unit 224 of the arrangement 700 may in this case be further configured to divert objects that is conveyed through the detection zone 720 based on the detected characteristics of the objects. In other words, characteristics of objects may thus be determined by the artificial neural network from the, by the camera, acquired images. The characteristics may be a shape, a color, features at the surface or anything in the visual appearance of the objects that may be determined and classified by the artificial neural network. The camera 750 may provide the possibility to further sort objects into different fractions. With the help of the artificial neural network, it may be possible to sort objects of the same material composition into different fractions depending on quality and origin.

[0128] Also depicted in Fig. 4 is a monitoring arrangement 20. The monitoring arrangement comprises a sensor arrangement, such as an RGB camera, to acquire monitoring sensor data of the objects being transported. The monitoring arrangement 20 may be configured to classify objects based on at least one property indicated for objects indicated by the monitoring sensor data.

[0129] The sensor arrangement of the monitoring arrangement and / or the sorting arrangement may comprise at least one sensor arrangement selected from the group comprising or consisting of: a visible spectrum, VIS, camera; a near-infrared, NIR, sensor; an infrared, IR, sensor; a hyperspectral imaging sensor; a thermal infrared sensor; a 3D stereo vision system; a lidar sensor; a time-of-flight, ToF, sensor; a laser scanner; an X-ray sensor; a polarimetric sensor, laser-induced breakdown spectroscopy, X-ray fluorescence, XRF; a temperature sensor; a humidity sensor; a vibration sensor; a gas sensor; a pressure sensor; a magnetic sensor; an electromagnetic sensor; an electrochemical sensor; a flow sensor; a proximity sensor; an ultrasonic sensor; an acoustic sensor.

[0130] Fig. 5 shows a graph over precision of statistics provided by the method according to some embodiments. In some embodiments, the data acquisition function is configured to selectively switch, over time, between different subsets of the at least one monitoring arrangement and the plurality of sorting arrangements from where to acquire sensor data from. For instance, referring to Fig. 3, sensor data may be received from each monitoring arrangement 20a, 20b during a first time interval and sensor data may be received from each sorting arrangement 30a, 30b during a second time interval. The data acquisition function may switch from which of these two subsets from where to acquire sensor data from. This may be referred to as multiplexed analysis. By this, the computational burden may be reduced, at the expense of foregoing constant analysis from each source. However, depending on the situation and overall design of the processing plant, it may be advantageous to implement multiplexed analysis, in particular for very large processing plants with many sources of sensor data. The data acquisition function may for instance monitor sensor data from each subset of sensor-equipped arrangements for a predetermined or configurable time period, and cycle through each subset of sensor-equipped arrangements. Fig. 5 shows a graph illustrating the impact of multiplexed analysis in comparison with real-time value (measured constantly). The thick solid line indicates a real value of a monitored parameter, such as a sorting quality of a sorted fraction, which for instance may be 99% or more, depending on situation and configuration of the processing plant. The thin solid line corresponds to a real-time value for the real value based on constant monitoring. The dashed line corresponds to a multiplexed value for the real value based on multiplex analysis, which involves intermittent monitoring. For short term statistics, such as on a time scale of minutes, they may differ due to sampling effects. However, over time, the results should average out to be equal. Time to may indicate a time where the multiplexed value can be similar to the real value, or at least be as accurate as an average value as the real-time value.

[0131] The plant-level monitoring of the processing plant may comprise, or be assisted by, a monitoring method, wherein the monitoring method comprises determining a holistic efficiency of a sorting process of at least one sorting system of the processing plant. The monitoring method comprises a step of determining the holistic efficiency £■(%) based at least on a plurality of efficiency parameters. The holistic efficiency £"(%) may be determined as wherein the parameters Av, Pf, I, R, Pt, and Q are efficiency parameters and represent equipment availability, feed performance, incoming target material, recovery rate, purity, and quality, respectively. The holistic efficiency E(%) may be based on a subset of these efficiency parameters. The holistic efficiency E(%) may be based on these efficiency parameters in combination with at least one further efficiency parameter. The data for determining the efficiency parameters may be obtained by a monitoring system, such as the monitoring system herein disclosed. The monitoring system may be configured to monitor one or more aspects of the at least one sorting system and / or one or more aspects of incoming feed material and / or one or more aspects of sorted, for instance via at least one sensor-equipped arrangement as described herein. The monitoring system may receive or access information about anyone of these aspects.

[0132] The equipment availability, represented by efficiency parameter Av, may be determined as a ratio of an actual equipment operating time of the at least one sorting system to a scheduled operating time of the at least one sorting system. A scheduled operating time may be an X amount of time per time unit, such as per day, per week, per month, etc. The equipment availability may be represented as a percentage. If the at least one sorting system comprises a plurality of sorting systems, the equipment availability Av may be an average or a mean of the equipment availability of each sorting system. The equipment availability Av may be an average or a mean of the equipment availability of the at least one sorting system over time. As a non-limiting example, the at least one sorting system is one sorting system, and the sorting system has an equipment operating time of 30 minutes in a hour (caused by a sorting system stop) and a scheduled operating time of 60 minutes for that hour, thereby yielding an equipment availability Av of 50% for that hour. The actual equipment operating time and the scheduled operating time of a sorting system or other equipment may be obtained or provided by the monitoring system.

[0133] The feed performance, represented by efficiency parameter Pf , may be determined as a ratio of total feed material received by the at least one sorting system to a total feed material capacity of the at least one sorting system. The feed performance may be represented as a percentage. If the at least one sorting system comprises a plurality of sorting systems, the feed performance Pf may be an average or a mean of the feed performance of each sorting system. The feed performance Pf may be an average or mean of the feed performance of the at least one sorting system over time. As a non-limiting example, the at least one sorting system is one sorting system, and the sorting system receives a total of 15 metric tons of feed material in an hour and has a total feed material capacity of 25 metric tons of feed material per hour, thereby yielding a feed performance of 60% for that hour. The total feed material received by the at least one sorting system and the total feed material capacity of the at least one sorting system may be obtained or provided by the monitoring system.

[0134] The incoming target material, represented by efficiency parameter / , may be determined as a ratio of how much material is expected to be recovered in the feed material to the total feed material. The incoming target material may be represented as a percentage. If the at least one sorting system comprises a plurality of sorting systems, the incoming target material may be an average or a mean of the incoming target material of each sorting system. The incoming target material I may be an average or mean of the incoming target material I of the at least one sorting system over time. As a non-limiting example, the at least one sorting system is one sorting system, and the sorting system receives a total of 15 metric tons of feed material in an hour, and the expected amount of target feed material to be recovered in the total feed material is 5 metric tons, thereby yielding an incoming target material I of 30% for that hour. The expected amount of feed material to be recovered by the at least one sorting system and the total feed material received may be obtained or provided by the monitoring system.

[0135] The recovery rate, represented by efficiency parameter R, may be determined as a ratio of how much material was actually recovered from the total amount of incoming target material. The recovery rate may be represented as a percentage. If the at least one sorting system comprises a plurality of sorting systems, the recovery rate R may be an average or a mean of the recovery rate R of each sorting system or of each target feed material. The recovery rate R may be an average or mean of the recovery rate R of the at least one sorting system over time. As a non-limiting example, the at least one sorting system is one sorting system, and the sorting system is expected to recover 5 metric tons of a target feed material from total feed material in an hour and actually recovers 4 metric tons of the target feed material from the total feed material in the hour, thereby yielding a recovery rate R of 80% for that hour. The amount of recovered target feed material and total amount of incoming target material may be obtained or provided by the monitoring system.

[0136] The purity, represented by efficiency parameter Pt, may be determined as a ratio of the amount of correctly sorted feed materials to a total amount of sorted feed materials. The purity may be represented as a percentage. The purity may be an average or a mean of the purity of sorted feed material. The purity may be an average or a mean of the purity of sorted feed materials over time. As a non-limiting example, the at least one sorting system is one sorting system, and the sorting system sorts a target feed material into a first material stream, wherein the first material stream has a contamination of 0.4 ton of erroneous sorted feed material in an hour and the total sorted feed materials in the first material stream for that hour is 4 tons, thereby yielding a purity of 90% for that hour. The amount of correctly sorted feed material to total amount of sorted feed materials may be obtained or provided by the monitoring system.

[0137] The quality, represented by efficiency parameter Q, may be determined as a ratio of an amount of bales of material that meet specifications. In material recovery and processing plants, incoming feed material is received in the at least one sorting system, typically in a loose, uncompressed form. The at least one sorting system separates the feed material into defined categories using mechanical and / or manual techniques based on analysis of the feed material. Each sorted material stream may then be directed to a baling apparatus, where material of each stream is compacted and bound to form bales suitable for storage, handling, and transport. A bale typically has a standardized size and shape. Each bale can be characterized in terms of specification, i.e., whether the material contents of the bale satisfy certain quality, composition, and / or physical requirements set by a buyer, industry standard, and / or a regulatory body. Thus, the at least one sorting system may produce bales as the final output of the sorting process. The quality may be represented as a percentage. The quality may be an average or a mean of the quality of a plurality of bales of a particular target feed material. The quality may be an average or a mean of the quality of a plurality of bales over time. As a non-limiting example, the at least one sorting system comprises one sorting system, and the sorting system produces a first material stream which is directed to a baling apparatus, which compacts material in the first material stream into 8 bales, each bale having a weight of 500 kg, and wherein 6 of the bales have a contamination of less than 10% of erroneous sorted material and 2 bales have a contamination of more than 10% of erroneous sorted material, and the specification for the bales specifies less than 10% of erroneously sorted materials, thereby yielding a quality of 75%. The contents of a bale and the specification for the bale may be obtained or provided by the monitoring system.

[0138] Using the values in the plurality of non-limiting examples above for the respective efficiency parameters, the holistic efficiency E becomes:

[0139] The holistic efficiency E advantageously enables characterization of sorting efficiency of the at least one sorting systems. The low holistic efficiency of the nonlimiting example above can partly be attributed to the relatively low equipment availability, feed performance, and incoming target feed material. Addressing these aspects may result in an improvement of the holistic efficiency. Further, it is preferable that the holistic efficiency E is maximized.

[0140] The monitoring method may comprise adjusting at least one system parameter of at least one sorting system of the at least one sorting system to increase a value of at least one efficiency parameter. System parameters of the at least one sorting system may be adjusted so as to increase, preferably maximize, each parameter. As a nonlimiting example, a system parameter may be a conveyor speed of a conveyor system associated with a sorting system, wherein the conveyor system transports feed material to the sorting system for sorting, wherein the conveyor speed of the conveyor system is increased, thereby increasing a throughput of feed material for sorting by the sorting system, yielding an increased feed performance Pf.

[0141] However, two or more of the efficiency parameters may be interdependent, so that adjusting a system parameter so that a value of a first efficiency parameter increases whereas a value of a second efficiency parameter decreases. Depending on how the values of the first efficiency parameter and the second efficiency parameter changes, the change in the system parameter may negatively impact the holistic efficiency despite attempting to improve it by adjusting the system parameter so that the value of the first efficiency parameter increases. As a non-limiting example, when increasing the feed performance, a sorting system has to sort feed material faster, which may negatively impact a sorting quality and consequently result in a reduced purity P. In other words, when adjusting a system parameter of at least one sorting system to increase a value of a first efficiency parameter, a value of a second efficiency parameter may be negatively affected by the change in the system parameter.

[0142] The monitoring method may comprise determining which system parameter or system parameters to adjust so that the efficiency parameters changes in such a way so that the holistic efficiency increases and / or is maintained equal to or greater than a minimum holistic efficiency threshold. The monitoring method may comprise simulating a sorting process of the at least sorting system, each simulation having a different set of system parameters for the at least one sorting system, and determining values of the efficiency parameters for the different set of system parameters, and based at least on an analysis of how the holistic efficiency changes with respect to the system parameters, determining a set of system parameters for the at least one sorting system corresponding to an increase in holistic efficiency and / or corresponding to the holistic efficiency being maintained equal to or greater than the minimum holistic efficiency threshold.

[0143] The monitoring method may iteratively adjust at least one system parameter to improve the holistic efficiency or maintain the holistic efficiency being equal to or greater than a minimum threshold. Each iteration, or cycle, may involve a process comprising: checking availability of an equipment, such as a sorting system; stabilizing feed performance; observing incoming feed material, and enforcing purity and recovery.

[0144] In connection with the step of checking availability of equipment, the monitoring system may determine whether an equipment is available for sorting. An equipment may be assigned with an operating state, which may indicate the equipment being available or unavailable. If available, the equipment is included for system optimization to improve or maintain a holistic efficiency. If unavailable, the equipment is not included for the system optimization. If the unavailability is unscheduled, the monitoring system may trigger an alarm to indicate that the equipment is faulty.

[0145] In connection with the step of stabilizing feed performance, the monitoring system may measure real-time flow of feed material by means of at least one sensor and compare the measured real-time flow of feed material with a target real-time flow of feed material, wherein the target real-time flow of feed material is based at least on available capacity. If the real-time flow of feed material deviates by a certain level from the target real-time flow of feed material, the monitoring system may stabilize feed performance. In connection with the step of observing incoming feed material, the monitoring system may repeatedly and / or continuously sample a feed composition of the incoming feed material by means of at least one sensor, and detect a change in feed composition based on an analysis of the sampled feed composition. If a feed composition remains at least one average within a target range, e.g., ±2% of a target feed composition, for a certain time duration, e.g., 5 min, the sorting process may proceed accordingly. If a change of feed composition outside the target range persists for a minimum time duration, e.g., 2 min, the monitoring system determines a new sorting configuration, each sorting configuration being simulated for current feed compositions, and then rank each sorting configuration according to their projected holistic efficiencies, and adjust the at least one sorting system based on the sorting configuration having the highest projected holistic efficiency. Alternatively, or in combination, in connection with the step of observing incoming feed material, the monitoring system may iteratively simulate a new sorting configuration for current feed compositions, and determine a holistic efficiency for each simulated sorting configuration, and if a new sorting configuration yields a holistic efficiency at least a minimum threshold larger than a current holistic efficiency without breaching other sorting constraints, adjust the at least one sorting system to the new sorting configuration.

[0146] In connection with the step of enforcing purity and quality, the monitoring system may monitor purity of a target feed material, and if purity is below a minimum purity threshold, the monitoring system determines a new sorting configuration based on the aforementioned simulation, and if a new sorting configuration maintains a recovery is equal to or larger than a minimum recovery threshold, adjust the at least one sorting system to the new sorting configuration. After achieving a desired purity, the monitoring system may gradually (e.g., stepwise) adjust system parameters to drive higher recovery until a plateau in holistic efficiency is reached or until purity reaches the minimum purity threshold.

[0147] The monitoring system may repeat the entire cycle a plurality of times. The entire cycle may be repeated at a fixed or adaptive period (e.g., every 30 seconds). The monitoring system may, for each cycle, scan for at least one high-priority condition. For instance, a high-priority condition may be functioning equipment or no presence of hazards in the feed material. If at least one high-priority condition is violated, such as an equipment is faulty, or there is a hazard in the feed material, the monitoring system adjusts at least one system parameter to address the violated high-priority condition. Once the high-priority condition is resolved, the monitoring system resumes optimization of a subsequent priority metric, for instance, holistic efficiency. According to a non-limiting example, the monitoring system iteratively optimizes holistic efficiency using a layer-based optimization, wherein the monitoring system only optimizes a subsequent layer if a precursor layer is optimized to a target optimization. The layers may comprise layers corresponding to the aforementioned steps of the cycling process, namely: a first layer of checking availability of an equipment; a second layer of stabilizing feed performance; a third layer of observing incoming feed material, and a fourth layer of enforcing purity and recovery, wherein the first layer is of highest priority and each subsequent layer being of successively lower priority.

[0148] In a non-limiting example, in a first cycle 1 , the monitoring system determines that a sorting system is online. The sorting system receives a flow of feed material with a mass flow rate of 21 tons / h (within target range of 5% of target 20 tons / h). The target material composition in the feed material is stable. The purity is 92% and above a minimum purity threshold of 90%. The holistic efficiency is determined as 46%, and as all constraints are satisfied for this sorting process, the monitoring system takes no action. In a second cycle 2, a sudden change in feed material is observed, the change being a reduction of target material from 75% to 68%. Purity drops to 88%. As purity drops to below the minimum purity threshold of 90%, the monitoring system initiates a process of determining a new sorting configuration by simulation. After simulating three new sorting configurations, the monitoring system determines that only one of these new sorting configurations achieves a purity above the minimum purity threshold of 90%, namely a purity of 91 %, for which the recovery is 98%. The monitoring system adjusts the sorting system to the new sorting configuration, and the holistic efficiency is increased to 51 %. In cycle 3, the monitoring system detects a mass flow rate of 22 tons / h for the flow of feed material received by the sorting system, which mass flow rate exceeds an upper limit of the target range for the mass flow rate. The monitoring system initiates a process of determining a new sorting configuration by simulation. After simulating three new sorting configurations, the monitoring system determines that only one of these new sorting configurations achieves a mass flow rate within the target range, the new system configuration comprising a reduced conveyor speed of a conveyor system feeding feed material to the sorting system, wherein the conveyor speed is reduced by 5%, for which the mass flow rate is returned to within the target range for the mass flow rate. The associated holistic efficiency for the new system configuration is 55%. As all constraints are satisfied for this sorting process once more, the monitoring system takes no action.

[0149] Notwithstanding the interdependency of efficiency parameters, it is typically desirable to increase each efficiency parameter. Such an increase may involve a plurality of aspects such as proactive and reactive action for the at least one sorting system, locally and / or remotely, and analysis of equipment to detect areas of improvements. For instance, a plant operator may perform tasks locally at the at least one sorting system. For instance, a process engineer may oversee quality of execution of tasks locally and / or remotely.

[0150] An increase in equipment availability may be promoted as follows. Alarms may be implemented to prevent or respond to sorting system failure or failures of other equipment. Alarms may be implemented to warn about degrading equipment health. Maintenance scheduling may further promote equipment availability, so that equipment does not experience sudden faults. In addition, equipment may be analyzed in terms of operating time, maintenance events, stops and failures, intervention history. The analysis may indicate how operation of the equipment may be improved, and documentation may be provided so as to train plant operators for improved operation of equipment.

[0151] An increase in feed performance may be promoted as follows. Plant operators and / or monitoring system may react to alarms of feed performance changing to outside a target range. Analysis of feed performance may detect areas of improvement. For instance, variability of feed material may be analyzed, and system configurations may be implemented to reduce variability of feed material.

[0152] An increase in parameter value for efficiency parameter of incoming target material may be promoted as follows. Alarms may be implemented to indicate when composition of feed material change compared to agreed thresholds. If composition of feed material changes, a more appropriate sorting program may be assumed. Any change in sorting program may be supervised by monitoring system. Any change in composition may be documented, and a supplier of the feed material may be formed accordingly of the change in composition.

[0153] An increase in recovery, quality, and / or purity may be promoted as follows. Sorting program may be changed to better suit feed material being received. Alarms may be implemented to indicate areas of improvement relating to the sorting process of the sorting system. Analysis of the sorting process may detect aspects of the sorting process, such as selection of system parameters in a given sorting configuration to improve recovery, quality and / or purity. Analysis may also indicate anyone sorting system having a suboptimal sorting configuration, and a more suitable sorting configuration may be implemented instead. While the foregoing is directed to embodiments of the disclosure, other and further embodiments may be devised without parting from the inventive concept discussed herein. The scope of the invention is however determined by the claims.

Claims

48CLAIMS1. A processing plant (1000), comprising: at least one sorting system (100-1 , 100-NP), each sorting system comprising: a transport arrangement (10) configured to transport objects as at least one stream of objects through a plurality of detection zones (DZ), the plurality of detection zones comprising a first detection zone and at least one subsequent detection zone arranged downstream of the first detection zone; at least one monitoring arrangement configured to acquire monitoring sensor data of objects when present in a predetermined one of the plurality of detection zones, and classify objects with monitoring object classification data; a plurality of sorting arrangements, wherein the plurality of sorting arrangements comprises Nssorting arrangements, wherein Ns> 10, wherein the plurality of sorting arrangements comprises at least one upstream sorting arrangement and at least one set of Npparallel sorting arrangements configured to directly receive respective fractions from the same upstream sorting arrangement, wherein Np> 4, wherein each sorting arrangement comprises an object displacement device, each sorting arrangement configured to: acquire sorting sensor data of objects when present in a predetermined one of the plurality of detection zones; provide sorting object classification data indicative of at least one object property of the objects based on the sorting sensor data; provide a sorting instruction indicative of at least one of at least two fractions based on said sorting object classification data, and sort the objects by means of the object displacement device into at least two fractions based on the sorting instruction, wherein, of the at least one monitoring arrangement and the plurality of sorting arrangements, only the plurality of sorting49 arrangements are configured to provide sorting instructions indicative of the at least one fraction to the object displacement devices; at least one processing circuitry (40), wherein the at least one processing circuitry (40) is configured to execute a plurality of functions, wherein said plurality of functions comprises: a data acquisition function configured to receive sensor data comprising the monitoring sensor data and the sorting sensor data, and process the sensor data to provide unified sensor data; a classification function configured to analyze the unified sensor data by at least one classifier based on at least one classification criterion, and generate at least one classification output for the unified sensor data, and a control function configured to provide output data based on the at least one classification output, wherein the output data comprises at least one of: control data for controlling the processing plant; status data indicative of a status of the processing plant, and property data indicative of at least one property of transported objects.

2. The processing plant (1000) according to any one of the preceding claims, wherein the control data comprises: data configured to initiate a control of at least one device and / or system of the processing plant and / or data configured to initiate a safety response of the processing plant.

3. The processing plant (1000) according to any one of the preceding claims, wherein the status data comprises data relating to at least one of: an overall throughput of the processing plant (1000); throughput within any one detection zone (DZ) of the plurality of detection zones (DZ); a status of at least one device of the processing plant; a layout change of the processing plant (1000); a blockage within any one detection zone (DZ) of the plurality of detection zones (DZ); a fire hazard within any one detection zone (DZ) of the plurality of detection zones (DZ), and50 a fault of any one device of the processing plant (1000).

4. The processing plant (1000) according to any one of the preceding claims, wherein the property data comprises data indicative of at least one of: at least one property of the objects being transported and / or sorted; at least one statistics of the objects being transported and / or sorted; a sorting quality of at least one sorting arrangement (30) or of at least one sorting system; accumulative sorting information during a predetermined time period; evenness of flow of objects within any one detection zone (DZ) of the plurality of detection zones (DZ).

5. The processing plant (1000) according to any one of the preceding claims, configured to transmit the output data to a display device and / or a communication device configured to present information based on the output data to an operator of the processing plant.

6. The processing plant (1000) according to any one of the preceding claims, wherein said process of the sensor data to provide unified sensor data comprises at least one of: normalizing and / or scaling the values of the received sensor data into new values, which new values are different from the values of the received sensor data; converting the values of the received sensor data into new values according to a frame of reference, wherein the frame of reference for the new values is different from the frame of reference of the values of the received sensor data; providing at least one semantic interpretation to the received sensor data, which at least one semantic interpretation is not present in the received sensor data; categorizing and optionally order the values of the received sensor data into new values according to a predefined system, such as by detection zones or sensor types normalizing, scaling and / or converting image data preferably with respect51 to resolution and / or spatial selection; formatting the received sensor data into new sensor data using a structural format different from that of the original sensor data, such as through reformatting, reordering, removal, or nesting of data elements; correlating the received sensor data into new sensor data, which new sensor data is correlated and / or synchronized in time or space, combining the received sensor data into new sensor data by merging and / or fusing sensor data of e.g. different sensor types.

7. The processing plant (1000) according to any one of the preceding claims, wherein each classifier of the at least one classifier comprises a machine learning model, optionally a neural network or a deep neural network, wherein each classifier is trained on labeled training data to monitor a respective aspect of the at least one aspect of the processing plant (1000).

8. The processing plant (1000) according to any one of the preceding claims, wherein the at least one classifier comprises a first classifier and at least a second classifier, wherein the first classifier is configured to: analyze the unified sensor data based on a first classification criterion, and generate a first classification output for the unified sensor data indicative of a first aspect of the processing plant (1000), and the second classifier is configured to: analyze image data based on a second classification criterion, and generate a second classification output for the unified sensor data indicative of a second aspect of the processing plant (1000).

9. The processing plant (1000) according to any one of the preceding claims, wherein at least one of the at least one monitoring arrangement and the plurality of sorting arrangements comprises at least one sensor arrangement selected from the group comprising or consisting of: a visible spectrum, VIS, camera; a near-infrared, NIR, sensor; an infrared, IR, sensor; a hyperspectral imaging sensor; a thermal infraredsensor; a 3D stereo vision system; a lidar sensor; a time-of-flight, ToF, sensor; a laser scanner; an X-ray sensor; a polarimetric sensor, laser-induced breakdown spectroscopy, X-ray fluorescence, XRF; a temperature sensor; a humidity sensor; a vibration sensor; a gas sensor; a pressure sensor; a magnetic sensor; an electromagnetic sensor; an electrochemical sensor; a flow sensor; a proximity sensor; an ultrasonic sensor; an acoustic sensor.

10. The processing plant (1000) according to any one of the preceding claims, wherein the at least one monitoring arrangement and the plurality of sorting arrangements each comprises at least one sensor arrangement of the same type.11 . The processing plant (1000) according to any one of the preceding claims, wherein the data acquisition function is configured to selectively switch, over time, between different subsets of the at least one monitoring arrangement and the plurality of sorting arrangements from where to acquire sensor data from.

12. The processing plant (1000) according to claim 12, wherein the different subsets of the at least one monitoring arrangement and the plurality of sorting arrangements comprises: a first subset and at least a second subset, wherein each of the at least one monitoring arrangement and the plurality of sorting arrangements of the first subset and the second subset is configured with a sensor arrangement of a first sensor type; and / or wherein each of the at least one monitoring arrangement and the plurality of sorting arrangements of the first subset is configured with a sensor arrangement of a first sensor type and each of the at least one monitoring arrangement and the plurality of sorting arrangements of the second subset is configured with a sensor arrangement of a second sensor type.

13. The processing plant (1000) according to any one of the preceding claims, wherein at least one sorting system of the plurality of sorting arrangements comprises: at least one processing circuitry (40) of the at least one processing circuitry (40).

14. The processing plant (1000) according to any one of the preceding claims, wherein at least one monitoring arrangement of the at least one monitoring arrangement comprises: at least one processing circuitry (40) of the at least one processing circuitry (40).

15. The processing plant (1000) according to any one of the preceding claims, wherein the at least one processing circuitry (40) comprises a remote processing circuitry, such as a cloud server, configured to execute at least one of the plurality of functions, and / or the at least one processing circuitry (40) comprises a local processing circuitry, such as a microcontroller or a computer arranged at a processing facility of the processing plant (1000), configured to execute at least one of the plurality of functions, and / or the at least one processing circuitry (40) comprises a portable processing device, such as a smart device, configured to execute at least one of the plurality of functions.

16. The processing plant (1000) according to any one of the preceding claims, wherein the at least one processing circuitry (40) comprises a plurality of processing circuitries configured to be communicatively connected, and wherein a first processing circuitry of the plurality of processing circuitries is configured to execute at least one function of the plurality of functions and at least a second processing circuitry is configured to execute at least one other function of the plurality of functions.5417. The processing plant (1000) according to any one of the preceding claims, wherein the at least one processing circuitry (40) is configured to execute at least one of the plurality of functions in real time.

18. The processing plant (1000) according to any one of the preceding claims, wherein the unified sensor data comprises image data comprising a two- dimensional array of pixels, each pixel having an associated intensity and / or color value and / or a corresponding geometry transformation mapping the pixel to a location or direction in a predetermined reference system.

19. The processing plant (1000) according to any one of the preceding claims, wherein the unified sensor data comprises non-image data, such as a time series of a measured parameter, or metadata defining at least one of: spatial resolution, color space, sensor type, sensor identification information, sensor system location, timestamps, calibration information, and environmental information.

20. The processing plant (1000) according to any one of the preceding claims, wherein the plurality of functions comprises: a data logging function configured to store the image data, the sensor data, classification output of the at least one classifier, and / or a control action for the processing plant.

21. The processing plant (1000) according to any one of the preceding claims, comprising at least one communication interface, each communication interface configured to communicatively connect, by wired connection and / or by wireless connection, at least one sorting arrangement of the plurality of sorting arrangements to the at least one processing circuitry (40).

22. The processing plant (1000) according to any one of the preceding claims, comprising a plurality of sorting assemblies, where each sorting system is configured to operate independently of other at least one other sorting system of the plurality of sorting assemblies.5523. The processing plant (1000) according to any one of the preceding claims, wherein the predetermined configuration of at least one set of sorting arrangement is configured to operate with at least one of a linear configuration, a branching configuration, and a merging configuration.

24. A monitoring system for a processing plant (1000), wherein the processing plant (1000) comprises: at least one sorting system, each sorting system comprising: a transport arrangement (10) configured to transport objects as at least one stream of objects through a plurality of detection zones (DZ), the plurality of detection zones comprising a first detection zone and at least one subsequent detection zone arranged downstream of the first detection zone; at least one monitoring arrangement configured to acquire monitoring data of objects when present in a predetermined one of the plurality of detection zones, and classify objects with at least one monitoring object classification; a plurality of sorting arrangements, wherein the plurality of sorting arrangements comprises Nssorting arrangements, wherein Ns> 10, wherein the plurality of sorting arrangements comprises at least one upstream sorting arrangement and at least one set of Npparallel sorting arrangements configured to directly or indirectly receive respective fractions from the same upstream sorting arrangement, wherein Np> 4, wherein each sorting arrangement comprises an object displacement device, each sorting arrangement configured to: acquire sorting sensor data of objects when present in a predetermined one of the plurality of detection zones; provide sorting object classification data indicative of at least one object property of the objects based on the sorting sensor data; provide a sorting instruction indicative of at least one of at least two fractions based on said sorting object classification data, and56 sort the objects by means of the object displacement device into at least two fractions based on the sorting instruction, wherein, of the at least one monitoring arrangement and the plurality of sorting arrangements, only the plurality of sorting arrangements are configured to provide sorting instructions indicative of the at least one fraction to the object displacement devices; wherein the monitoring system comprises: at least one processing circuitry (40), wherein the at least one processing circuitry (40) is configured to execute a plurality of functions, wherein said plurality of functions comprises: a data acquisition function configured to receive sensor data comprising the monitoring sensor data and the sorting sensor data, and process the sensor data to provide unified sensor data; a classification function configured to analyze the unified sensor data by at least one classifier based on at least one classification criterion, and generate at least one classification output for the unified sensor data, and a control function configured to provide output data based on the at least one classification output, wherein the output data comprises at least one of: control data for controlling the processing plant; status data indicative of a status of the processing plant, and property data indicative of at least one property of transported objects.

25. The monitoring system according to claim 25, wherein the supervisory system is configured as a supervisory control and data acquisition, SCADA, system.

26. Method of monitoring a processing plant (1000), wherein the processing plant comprises: at least one sorting system (100-1 , 100-NP), each sorting system comprising: a transport arrangement (10) configured to transport objects as at least one stream of objects through a plurality of detection zones (DZ), the plurality of detection zones comprising a first detection zone and at57 least one subsequent detection zone arranged downstream of the first detection zone; at least one monitoring arrangement configured to acquire monitoring data of objects when present in a predetermined one of the plurality of detection zones, and classify objects with monitoring object classification data; a plurality of sorting arrangements, wherein the plurality of sorting arrangements comprises Nssorting arrangements, wherein Ns> 10, wherein the plurality of sorting arrangements comprises at least one upstream sorting arrangement and at least one set of Npparallel sorting arrangements configured to directly or indirectly receive respective fractions from the same upstream sorting arrangement, wherein Np> 4, wherein each sorting arrangement comprises an object displacement device, each sorting arrangement configured to: acquire sorting sensor data of objects when present in a predetermined one of the plurality of detection zones; provide sorting object classification data indicative of at least one object property of the objects based on the sorting sensor data; provide a sorting instruction indicative of at least one of at least two fractions based on said sorting object classification data, and sort the objects by means of the object displacement device into at least two fractions based on the sorting instruction, wherein, of the at least one monitoring arrangement and the plurality of sorting arrangements, only the plurality of sorting arrangements are configured to provide sorting instructions indicative of the at least one fraction to the object displacement devices; the method comprising: receiving sensor data comprising the monitoring sensor data and the sorting sensor data, and processing the sensor data to provide unified sensor data;analyzing the unified sensor data by at least one classifier based on at least one classification criterion, and generating at least one classification output for the unified sensor data, and providing output data based on the at least one classification output, wherein the output data comprises at least one of: control data for controlling the processing plant; status data indicative of a status of the processing plant, and property data indicative of at least one property of transported objects.

Images