System for monitoring, analyzing and controlling the quantity and / or concentration of process fluids, as well as sensor gateway, mixing and dosing module and evaluation and control unit for use in such a system

Abstract

System (10) for monitoring, analyzing and controlling the quantity and / or concentration of process fluids in at least one tank (12) and / or container in industrial plants and / or agricultural plants and / or technical processes, comprising: - at least one mixing and dosing module (14) which is configured to calculate and automatically execute a required re-dosing based on at least one current concentration and / or fill level value; - at least one networked level sensor (16) and / or level switch; - a hybrid measurement architecture with at least one digital handheld measuring device (20) and at least one automatic sensor; - a sensor gateway (18); - an evaluation unit; and - a control unit.

Description

[0001] The invention relates to automation technology and fluid management, in particular a system for monitoring, analyzing, and controlling the quantity and / or concentration of process fluids in technical processes. Such process fluids can be, for example, cooling lubricants in the metalworking industry, chemicals, cleaning solutions, wastewater, or liquid fertilizers in agricultural plants.

[0002] Systems for the automatic management of process fluids, such as cooling lubricants in machine tools, are known from the prior art. Such systems typically include sensors for detecting the fill level in a reservoir. If the level falls below a threshold, a mixing process is automatically triggered, in which a concentrate is mixed with water and added to the reservoir to restore the fill level. This process is often controlled based on a predefined, fixed mixing ratio.

[0003] Other well-known approaches are based on purely manual measurement routines. Here, operators take samples of the process fluid, measure parameters such as concentration with handheld measuring devices, and initiate dosing measures based on these manually recorded values.

[0004] However, fully automated systems often prove prone to malfunctions in real-world industrial environments. Process fluids are frequently contaminated with particles, oils, or abrasion, which can lead to fouling of continuously measuring in-line or in-tank sensors. This results in inaccurate measurements, interruptions in normal operation, and high maintenance costs, significantly impacting the economic viability of such systems.

[0005] Systems based on manual measurement routines suffer from a lack of data integrity and traceability. Media breaks, such as when reading a measurement and manually entering it into a control system, lead to transmission errors. Furthermore, a verifiable, tamper-proof chain from measurement to dosing decision is lacking, which jeopardizes the reliability and safety of the entire process.

[0006] The present invention therefore aims to provide a system for monitoring, analyzing, and controlling process fluids that ensures high operational safety and reliability even under the harsh conditions of industrial environments with contaminated fluids, while simultaneously guaranteeing a complete, traceable, and tamper-proof data chain from measurement acquisition to dosing decision. A further aspect of the invention is the provision of a suitable sensor gateway, a mixing and dosing module, and an evaluation and control unit.

[0007] This problem is solved by a system having the features of claim 1, as well as by a sensor gateway according to claim 26, a mixing and dosing module according to claim 27, and an evaluation and control unit according to claim 28. Advantageous further developments are the subject of the dependent claims.

[0008] A system according to the invention for monitoring, analyzing, and controlling the quantity and / or concentration of process liquids in at least one tank or container comprises at least one mixing and dosing module, a networked level sensor, a hybrid measurement architecture with a digital handheld measuring device and an automatic sensor, a sensor gateway, an evaluation unit, and a control unit. The mixing and dosing module is configured to calculate and automatically execute a required replenishment based on a concentration and / or level value.

[0009] The decisive advantage of this system lies in the creation of a hybrid yet closed architecture. It resolves the fundamental conflict between fully automated systems, which fail in the realities of complex industrial environments, and purely manual processes, which are error-prone and unsafe. By combining an automatic sensor, typically for a robust measurement such as fill level, with a digital handheld measuring device for more demanding parameters like concentration, high measurement reliability is achieved for the overall system. Such a hybrid system can also be described as a "human-in-the-loop" system.

[0010] The digital nature of the handheld measuring device is of central importance here. Unlike analog devices, whose values ​​must be read and entered manually, the digital device enables direct, electronic transmission of the measured values ​​to the sensor gateway. This completely eliminates so-called media breaks. Sources of error such as reading errors, transposed digits, or decimal point errors during manual entry are inherently eliminated by the system. The basis for every control decision—the measured value—is thus of significantly higher quality and reliability from the outset. Alternatively, these operator tasks, in particular sampling, data acquisition, and feedback of the measured values, can also be performed semi- or fully automatically by a robot within a "robot-in-the-loop" approach. A combination of both approaches is also conceivable and can be understood as a "human & robot-in-the-loop" approach.

[0011] The sensor gateway acts as an intelligent hub and the heart of the data acquisition system. It is not merely a passive interface, but an active component that aggregates data from disparate sources – the automatic level sensor and the digital handheld measuring device. This aggregation ensures that the downstream evaluation and control unit receives a complete and coherent picture of the process fluid's condition. Without such a gateway, integrating various independent measurement paths into a single, closed control loop would be technically complex and prone to errors.

[0012] The entire system of components forms a seamless, closed-loop control loop. The acquired measured values ​​(actual values) are transmitted via the gateway to the evaluation and control unit, which processes the data, compares it with target values, and generates a control command for replenishment. This command is sent to the mixing and dosing module, which, acting as an actuator, adjusts the process fluid in the tank accordingly. This closed loop enables autonomous or semi-autonomous control, which maintains stable process quality and significantly reduces the manual effort required from operating personnel.

[0013] The system described above is not limited in its design to the aforementioned configuration of a single automatic sensor and a handheld measuring device. Rather, it is modular and allows for the flexible integration of additional automatic sensors. This makes it possible to implement full local automation in selected areas of a plant where, for example, continuous monitoring of concentration or oil content is required and technically feasible, while other areas benefit from the hybrid architecture. Such additional fully automatic in-line or in-tank sensors can be directly integrated into the system via standard analog or digital interfaces to ensure continuous, automatic data acquisition in these areas as well.

[0014] According to a preferred embodiment, the feedback of manually acquired measurement values ​​can be coupled to a unit with integrated short-range communication. This unit can, in particular, be a smart sensor or another field device such as a valve that itself has an integrated communication interface—for example, NFC, Bluetooth, RFID, or WLAN functionality. A concrete example of this is an advanced level sensor. In this case, the measurement data from the digital handheld measuring device is transmitted directly to this smart sensor by simply bringing it close to or holding it near the tank. From there, combined with the automatically acquired data, it is fed into the control loop. This further simplifies the process for the operator and creates a single, robust data point at the tank itself.

[0015] According to a preferred embodiment, the evaluation unit and the control unit are combined into a single evaluation and control unit. This integration reduces hardware complexity, thereby lowering system costs and minimizing physical space requirements. Furthermore, the logical processes of data evaluation and control calculation can be more closely coupled, resulting in lower latency and more efficient processing.

[0016] According to a preferred embodiment, the hybrid measurement architecture acquires measurements of at least one concentration, at least one pH value, at least one conductivity, and / or other fluid parameters. This gives the system high flexibility and significantly expands its range of applications. Instead of controlling only a single parameter such as concentration, the system can perform multi-parameter control. For example, it can simultaneously optimize the concentration and pH value of a cooling lubricant, resulting in significantly improved process stability, longer fluid service life, and higher manufacturing quality.

[0017] According to a preferred embodiment, the at least one automatic sensor is a semi-automatic or fully automatic in-tank or in-line sensor. Such sensors are particularly suitable for measuring points requiring continuous monitoring, such as concentration or oil content. This flexibility in the choice of sensor technology allows for optimal adaptation to the specific application. For simple level measurements in an open tank, a cost-effective in-tank sensor may suffice, while for critical processes in closed lines, a precise in-line sensor may be required. The system is therefore not limited to a specific sensor technology and can be configured according to technical and economic requirements.

[0018] According to a preferred embodiment, the evaluation unit is configured to process and / or store measured values ​​acquired by the hybrid measurement architecture and to make them available to the control unit for fluid control. The ability to store measured values ​​creates a valuable data history. This historical data forms the basis for further analyses, such as trend detection (e.g., a gradual decrease in pH value) or the diagnosis of process disturbances. In particular, such analyses for trend detection and process optimization can also be AI-supported to recognize complex relationships and make predictions. This functionality transforms the system from a purely reactive controller into an intelligent monitoring instrument.

[0019] According to a preferred embodiment, a continuous digital data chain with integrity checking exists from the hybrid measurement architecture to the control unit. Integrity checking, for example using checksums or hashes, ensures that the transmitted data has not been corrupted during transmission (e.g., by transmission errors). This increases the reliability of the system, as the control unit can be confident that it is making decisions based on correct and unaltered data, thus preventing incorrect dosing due to data corruption.

[0020] According to a preferred embodiment, the sensor gateway is configured to aggregate acquired measurement values, temporarily store them with a timestamp, and transmit them to the evaluation unit. Adding a timestamp directly during acquisition improves data quality. It enables precise chronological assignment of the measurement values, which is essential for trend analysis and correlation with other process events. The ability to temporarily store data increases the robustness of the system, as measurement values ​​are not lost even if the connection to the evaluation unit is temporarily interrupted.

[0021] According to a preferred embodiment, the sensor gateway is configured to assign device and operator IDs, as well as temperature metadata, to the acquired measurements. This enrichment of the raw data with metadata creates valuable context for each individual measurement. For example, assigning a device ID allows for tracking the calibration history of individual measuring instruments. The operator ID establishes accountability and traceability for manual measurements. Temperature metadata is crucial because many chemical-physical measurements (e.g., concentration using a refractometer) are temperature-dependent and require compensation to deliver accurate results.

[0022] According to a preferred embodiment, the hybrid measurement architecture, the evaluation unit, and the control unit are designed to detect highly contaminated process fluids. By using targeted, manual sampling with an external digital measuring device for the critical measurement in the contaminated fluid, the system avoids the problems of sensor contamination. The system thus remains fully functional even under conditions where conventional fully automated systems would have already failed.

[0023] According to a preferred embodiment, the system, with its hybrid control loop in which acquired measured values ​​and model-based interpolated and / or predicted values ​​form a control model, constitutes a particularly intelligent control system. This increases the robustness and efficiency of the system. For example, if a measured value fails or is implausible, the system can interpolate or predict a probable substitute value based on historical data and a learned model. This allows the control operation to continue without interruption, thus maximizing system availability.

[0024] According to a preferred embodiment, if a handheld measuring device or a sensor in the hybrid measurement architecture fails, substitute values ​​from historical data and / or model-based estimates are used. This constitutes an intelligent fallback mechanism. Instead of halting operation in the event of a sensor failure, the system can generate a reliable estimate based on the stored data history and trained models. This ensures high reliability and continuous process operation, even if individual hardware components are temporarily unavailable.

[0025] According to a preferred embodiment, the system applies a model-based fallback procedure when measured values ​​are missing or contradictory. This procedure is particularly sophisticated because it considers not only historical measured values ​​but also correlations such as temperature corrections and fill level profiles, and generates a substitute value, including a confidence score, using weighted linear regression or a trained regression model. The confidence score indicates the reliability of the estimated value and enables the control unit to make a differentiated decision. This is significantly more advanced than simple interpolation and leads to more precise and reliable dosing decisions in the event of an error.

[0026] According to a preferred embodiment, the evaluation unit provides a proactive measurement tasking module. This transforms the system from a purely passive, reactive controller into a proactive management system. Instead of waiting for fixed measurement intervals, the system can independently request a new measurement if, for example, it detects an accelerating trend or high uncertainty in its forecasting models. This optimizes measurement effort, as measurements are only taken when truly necessary, and simultaneously increases process reliability, since critical deviations are detected early.

[0027] According to a preferred embodiment, a multimodal plausibility check is provided. This check verifies the reliability of a measured value not only based on numerical limits, but also combines this with structured operator input or even image data. If an implausibility is detected, a test report is automatically generated and a controlled action is initiated. This utilizes the operator's human expertise as an additional data source and significantly increases the reliability of detection for complex problems.

[0028] According to a preferred embodiment, the mixing and dosing module executes a control command only if it has a valid cryptographic signature. This is a crucial security mechanism. It ensures that the dosing module responds exclusively to authentic and unaltered commands from an authorized evaluation unit or sensor gateway. Manipulated or forged control commands are rejected, preventing unauthorized dosing and potential damage to equipment or products.

[0029] According to a preferred embodiment, hardware-based device attestation is used, whereby only attested devices are permitted to transmit measured values ​​or accept control commands. This raises system security to an even higher level. Using technologies such as a Trusted Platform Module (TPM) or Secure Element, each hardware component (gateway, dosing module) can unambiguously authenticate itself within the network. This ensures that no unauthorized or compromised devices can be introduced into the control loop, thus protecting the integrity of the entire system from the hardware level upwards.

[0030] According to a preferred embodiment, control commands and / or measured values, including metadata, are cryptographically signed before transmission. Signing the data directly at the source (e.g., at the sensor gateway) protects data integrity and authenticity throughout the entire transmission path. Subsequent instances can verify at any time that the data originates from a legitimate source and has not been altered since its creation.

[0031] According to a preferred embodiment, control commands to the mixing and dosing module include a verification hash. This verification hash is derived from the most recently received, signed measurement data, the calculated dosage quantity, and a timestamp. The dosing module verifies this verification hash before execution. This establishes a direct cryptographic link between the measurement on which the decision is based and the executed command. This ensures that a control command is executed only based on a specific, current, and valid data set, thus preventing replay attacks or the execution of outdated commands.

[0032] According to a preferred embodiment, the system accepts firmware updates or configuration changes only after validation of a manufacturer signature and / or verification of rollback protection mechanisms. This protects the system from the installation of manipulated or malicious firmware. The rollback protection also prevents attackers from installing an older, potentially vulnerable firmware version. The integrity and security of the system software are thus ensured throughout its entire lifecycle.

[0033] According to a preferred embodiment, all network-based connections are secured by means of mutual TLS authentication with certificate verification. This ensures strong, encrypted, and mutually authenticated communication between all system components (gateway, evaluation unit, dosing module). Unlike a simple TLS connection, here not only does the server authenticate itself to the client, but the client also authenticates itself to the server. This prevents man-in-the-middle attacks and ensures that the components communicate exclusively with legitimate, known counterparts.

[0034] According to a preferred embodiment, the mixing and dosing module responds to a control command only if it is signed with a specific, periodically generated, and centrally managed token. This implements a dynamic authorization system. Short-lived tokens are used instead of static keys or passwords. This significantly increases security, as a compromised token is only valid for a limited time. Furthermore, central management allows for the immediate revocation of access rights when necessary.

[0035] According to a preferred embodiment, a local verification stage is provided in the sensor gateway and / or in the mixing and dosing module. This stage checks the integrity, signature, and consistency against recently stored data records and blocks execution in case of deviations. This creates decentralized security intelligence. Even if the connection to the central evaluation unit is interrupted, the decentralized components can still perform a basic security check and prevent inconsistent or potentially malicious actions, further increasing the robustness of the overall system.

[0036] According to a preferred embodiment, the system is configured to execute a fully autonomous sequence of actions, including the acquisition of measured values, their transmission, the calculation of the dosage quantity, and the execution of the dosage command without manual confirmation. This enables fully automated operation for routine tasks, maximizing efficiency and reducing the workload for personnel. The system can independently control the process as long as all measured values ​​are plausible and all safety checks are successful.

[0037] According to a preferred embodiment, all events are logged in a tamper-proof audit log. The use of tamper-proof structures ensures that the log data cannot be subsequently altered without detection. This creates an audit-proof documentation of all system activities, which is of great value for quality assurance, troubleshooting, and compliance in regulated industries.

[0038] According to the invention, a sensor gateway is provided for use in such a system. This gateway is configured to aggregate measurement data from the various sources of the hybrid measurement architecture, time-stamp it, enrich it with important metadata such as device or operator IDs, and transmit the processed data packets cryptographically signed via standardized protocols. This gateway represents the crucial link and the security anchor of the system. It enables modular integration and retrofitting of existing systems by encapsulating the complex and safety-critical task of data acquisition and preprocessing in a dedicated unit.By enriching the raw data with context and immediately cryptographically signing it, an indisputable, high-quality and tamper-proof data basis is created, which is essential for subsequent regulatory decisions, audits and process analyses, while at the same time ensuring high interoperability with higher-level IT systems.

[0039] A mixing and dosing module according to the invention is intended for use in one of the systems described above and is configured to subject incoming dosing commands to a rigorous security check before execution. In contrast to a conventional actuator, this module acts as an intelligent last line of defense by verifying the cryptographic signature of a command to check its authenticity, as well as an included check hash to ensure contextual correctness and timeliness. Only after successful verification of both criteria is the dosing command executed automatically. This decentralized verification logic makes the system extremely robust against manipulation, replay attacks, or transmission errors and implements a fail-safe feature in which no action is taken in case of doubt.This reliably prevents process-critical incorrect dosing and ensures the safety and integrity of the physical process directly at the actuator.

[0040] An evaluation and control unit according to the invention is designed as the central intelligence for one of the systems described above. It is configured to first verify the authenticity and integrity of the received signed measurement values ​​in order to establish a trustworthy data basis. Based on this, it calculates the required dosage quantity and generates a cryptographically secured dosage command that contains both a digital signature and a context-related verification hash. Furthermore, the unit can act proactively by independently triggering new measurement tasks when trends or model uncertainties are detected. Every single step of this decision-making process—from data verification and command generation to task triggering—is seamlessly logged in a tamper-proof audit log, ensuring complete traceability and auditability for audits and analyses.

[0041] The invention is explained below with reference to the accompanying drawings. These show: Fig. 1 an illustrated overview of a preferred embodiment of the system according to the invention; Fig. 2 an embodiment of a control loop with level measurement and local data transmission; Fig. 3. An embodiment of a control loop with manual measurement data acquisition and a cloud portal.

[0042] Fig. Figure 1 shows an illustrated overview of a preferred embodiment of the system according to the invention. The overall system 10 is designed to control a process fluid in a tank 12. A level sensor 16 is arranged on the tank 12, which continuously detects the fluid level. A tank valve 24 for controlling the fluid flow is also located on the tank 12. For unambiguous identification, the tank 12 is provided with a QR code 22. Instead of the QR code 22, other identification means, such as NFC tags, RFID tags, or other electronic identification tags, can also be used. The use of such an identification means ensures an error-free assignment of the manually recorded measured values ​​to the correct measuring location, in this case, the specific tank 12.

[0043] A mixing and dosing module 14 is connected to the tank 12 and configured to supply it with a mixture of a concentrate and a liquid, for example water, as needed. A sensor gateway 18 acts as a central communication and data acquisition unit. It receives data from various sensors and devices and communicates with higher-level systems.

[0044] A digital handheld measuring device 20 is provided for the manual recording of measurement parameters such as concentration or pH value. This device can communicate with a smartphone or other transmission device, for example via NFC (Near Field Communication), to transmit the recorded measurements digitally and without errors. The collected data, processed via the sensor gateway 18, is transmitted to a central evaluation and control instance, such as a server 26 in a cloud infrastructure. Software running on this server 26 analyzes the data, makes control decisions, and makes them available to the user via a graphical user interface on a laptop or smartphone.

[0045] Fig. Figure 2 shows an embodiment of a control loop with level measurement and local data transmission. This control loop primarily serves to regulate the fill level in a tank. The setpoint or target fill level is specified either via an online or cloud connection in step S1 or, alternatively, set locally via an on-site menu in step S2. In the mixing and dosing module, which acts as the controller, a target-actual comparison is performed in step S3. The deviation between the target value and the measured actual value is used to calculate the required refill volume.

[0046] Based on this calculation, the controller in step S4 activates the actuators, such as a pump and a tank valve, to initiate the re-dosing process. The controlled system in step S5 is the tank itself, whose fill level is affected by the re-dosing. In step S6, disturbances affect the tank, such as liquid leakage through components, evaporation, or withdrawal. In step S7, the fill level is measured, meaning the actual fill level is automatically recorded in real time by a level sensor. The determined actual value is transmitted back to the controller in step S8 via a direct connection, which can be implemented locally via cable or wirelessly. Optionally, an edge computer or an advanced gateway can be interposed in step S9 to preprocess, enrich, or optimize the measurement data before it is used in step S3 for the target / actual comparison.

[0047] Fig.Figure 3 shows an embodiment of a control loop with manual data acquisition and optional connection to a cloud portal. This control loop is specifically designed for controlling quality parameters such as concentration or pH value. The reference variable or setpoint for the parameter to be controlled is either specified externally, for example via the internet, in step S11 or configured locally on the system in step S12. In the mixing and dosing module, which acts as the controller, the setpoint-actual value comparison takes place in step S13, and a corresponding adjustment concentration or dosage quantity is calculated.

[0048] In step S14, the actuators, i.e., the mixer and the tank valve, are controlled to supply the calculated dosage quantity to the controlled system, i.e., the tank, in step S15. In step S16, disturbances such as contamination, evaporation, or chemical decomposition can affect the process. The actual value is recorded in step S17 by manual measurement using a digital handheld measuring device. The digital transmission of the measured values ​​and tank identification, for example, from the handheld measuring device, takes place in step S18 via interfaces such as NFC, Bluetooth, or QR code to a transmission device such as a smartphone in step S19.

[0049] From there, alternative transmission paths exist. According to one option, a direct local connection via cable or radio to a locally based intelligence, such as an edge computer or gateway shown in step S21, can be established in step S20. This intelligence processes the data. Alternatively, the data is transmitted via the internet in step S22 to an optional cloud server in step S23, which serves as a central data repository and analysis portal. AI software for trend analysis and optimization can then run there in step S24. The verified actual value resulting from these steps is finally reported back to the controller in step S13 to close the control loop. Reference symbol list 10 System 12 Tank 14 Mixing and dosing module 16 Level sensor 18 Sensor Gateway 20 handheld measuring device 22 QR codes for tank identification 24 Tank valve 26 Servers

Claims

[1] System (10) for monitoring, analyzing and controlling the quantity and / or concentration of process fluids in at least one tank (12) and / or container in industrial plants and / or agricultural plants and / or technical processes, comprising: - at least one mixing and dosing module (14) which is configured to calculate and automatically execute a required re-dosing based on at least one current concentration and / or fill level value; - at least one networked level sensor (16) and / or level switch; - a hybrid measurement architecture with at least one digital handheld measuring device (20) and at least one automatic sensor; - a sensor gateway (18); - an evaluation unit; and - a control unit. [2] System (10) according to claim 1, wherein at least one unit arranged on the tank (12) or on the container has an integrated data interface, in particular a short-range communication interface, which is configured to receive measured values ​​from the digital handheld measuring device (20) by spatially approaching or holding it in front of the device without contact and without manual data input. [3] System (10) according to claim 1 or 2, wherein the evaluation unit and the control unit are combined in one evaluation and control unit. [4] System (10) according to one of the preceding claims, wherein the hybrid measurement architecture acquires measured values ​​for at least one concentration, at least one pH value, at least one conductivity and / or other fluid parameters. [5] System (10) according to any of the preceding claims, wherein the at least one automatic sensor is a semi-automatic or a fully automatic in-tank sensor or in-line sensor. [6] System (10) according to one of the preceding claims, wherein the evaluation unit is configured to process and / or store measured values ​​acquired by the hybrid measurement architecture and to make them available to the control unit for fluid control. [7] System (10) according to claim 6, wherein a continuous digital data chain with integrity check exists from the hybrid measurement architecture to the control unit. [8] System (10) according to one of the preceding claims, wherein the sensor gateway (18) is configured to aggregate measured values ​​acquired by the hybrid measurement architecture, to temporarily store them with a time stamp and to transfer them to the evaluation unit. [9] System (10) according to one of the preceding claims, wherein the sensor gateway (18) is configured to assign device and operator IDs and temperature metadata to measured values ​​acquired by the hybrid measurement architecture. [10] System (10) according to one of the preceding claims, wherein the hybrid measurement architecture, the evaluation unit and the control unit are designed such that highly contaminated process fluids containing solid particles, oil droplets, foreign substances and / or abrasion can be detected without a continuously flowing measurement path. [11] System (10) according to one of the preceding claims, comprising a hybrid control loop, wherein in the hybrid control loop measured values ​​acquired by the hybrid measurement architecture and model-based interpolated and / or predicted values ​​form a control model. [12] System (10) according to claim 11, wherein in the event of failure of at least one handheld measuring device (10) and / or one sensor of the hybrid measurement architecture, substitute values ​​from historical data and / or model-based estimates are used. [13] System (10) according to one of the preceding claims, wherein the system is configured to apply a model-based fallback procedure in the event of missing and / or contradictory measured values ​​acquired by the hybrid measurement architecture, in which historical measured values, temperature corrections and level profile correlations are taken into account and a substitute value with confidence score for a calculation of the re-dosing is generated by means of weighted linear regression and / or by means of a trained regression model. [14] System (10) according to one of the preceding claims, wherein the evaluation unit provides a proactive measurement tasking module which is configured to automatically generate measurement tasks based on trends and / or model uncertainties and to transmit these to the sensor gateway (18) and / or to an operating device via a bidirectional interface. [15] System (10) according to one of the preceding claims, comprising a multimodal plausibility check, wherein the multimodal plausibility check is configured to combine numerical measured values, structured operator inputs and / or image data and, in case of implausibility, to automatically generate a test protocol and initiate a controlled measure. [16] System (10) according to one of the preceding claims, wherein the mixing and dosing module (14) is configured to execute at least one control command only if it has a valid cryptographic signature that can be verified using a certificate chain stored in the evaluation unit and / or the sensor gateway (18). [17] System (10) according to one of the preceding claims, with hardware-based device attestation, wherein only attested devices with validated identities and firmware hashes are authorized to transmit measured values ​​acquired by the hybrid measurement architecture and / or to accept control commands. [18] System (10) according to one of the preceding claims, wherein the system is configured to cryptographically sign control commands and / or measured values ​​acquired by the hybrid measurement architecture, including metadata, prior to transmission. [19] System (10) according to one of the preceding claims, wherein control commands to the mixing and dosing module (14) contain a verification hash derived from the last signed measurement data received by the hybrid measurement architecture, a dosing quantity calculated by the control unit and a timestamp, and wherein the mixing and dosing module (14) is configured to verify the verification hash before executing the control command. [20] System (10) according to any of the preceding claims, wherein the system (10) is configured to accept firmware updates and / or configuration changes only after validation of a manufacturer signature and / or after verification of rollback protection mechanisms. [21] System (10) according to one of the preceding claims, wherein all network-based connections between the sensor gateway (18), the evaluation unit and the mixing and dosing module (14) are secured by mutual TLS authentication with certificate verification. [22] System (10) according to one of the preceding claims, wherein the mixing and dosing module (14) is configured to respond to a control command only when it is signed with a specific token that is periodically generated and centrally managed. [23] System (10) according to one of the preceding claims, comprising a local verification stage in the sensor gateway (18) and / or in the mixing and dosing module (14), wherein the local verification stage is configured to check integrity, signature and consistency against stored data sets and, in case of deviations, to block execution and trigger an alarm. [24] System (10) according to any of the preceding claims, configured to perform the following steps without manual confirmation: - Acquisition of measured values ​​at at least one tank (12) and / or at least one container with local timestamping and plausibility check by the hybrid measurement architecture; - Transfer to the evaluation unit; - Calculation of a required dosage quantity in the control unit; and - Transmission of a signed dosing command to the mixing and dosing module (14) and execution of the dosing command. [25] System (10) according to claim 24, further configured to record all events in a tamper-detecting audit log. [26] Sensor gateway (18) for use in a system (10) according to any one of claims 1 to 25, configured for aggregating, timestamping, enriching metadata and transmitting signed measurement values ​​via standardized protocols. [27] Mixing and dosing module (14) for use in a system (10) according to any one of claims 1 to 25, configured for verifying cryptographic signatures and check hashes of incoming dosing commands and for automatically executing verified dosing commands. [28] Evaluation and control unit for use in a system (10) according to one of claims 2 to 25, configured for this purpose: - to receive and verify signed measurement values; - to calculate the required dosage amount; - to generate a signed dosing command with a verification hash; - To trigger measurement tasks; and to log each of the above steps in a tamper-detecting audit log.

Images