Patent Application
20250059101
In general, the present application relates to the field of an AI-enabled technology used in concrete-production systems. In particular, the invention centres on a concrete production system equipped with a proactive AI-powered control system.
Concrete is the most widely used construction material around the world. It is a composite material with a complex structure composed of water, fine aggregates, coarse aggregates, sand, chemical additives, and various chemical admixtures bonded together with a fluid cement (cement paste) that hardens (cures) over time. Cement normally comprises from 10 to 15 percent of the concrete mix, by weight. Through a process called hydration reaction, the cement and water react, harden, and bind the aggregates into a rock-like mass. This setting and hardening process continues from when cement is mixed with water and may continue for several months, meaning concrete gets harder over time. Portland cement is not a brand name, but the generic term for the type of cement used in virtually all concrete, just as stainless is a type of steel and sterling a type of silver. Therefore, there is no such thing as a cement sidewalk, or a cement mixer; the proper terms are concrete sidewalk and concrete mixer.
Cement paste is produced by a rapid process of hydration of clinker minerals, releasing large amounts of heat once the cement is mixed with water over a period of minutes, hours and days. Tricalcium silicate (C3S) is one of the main cementitious components of Portland cement, and its hydration reaction is represented by the following chemical equation:
2Ca3SiO5+6H2O→Ca3Si2O7.3H2O+3Ca(OH)22C3S+6H→C3S2H3+3CH (in cement nomenclature)
The products formed by the slow hydration reaction over several weeks are calcium silicate hydrate, known as C—S—H, and calcium hydroxide. Dicalcium silicate (C2S) hydrates much more slowly than C3S does, to form similar type of C—S—H and Ca(OH)2:
2Ca2SiO4+4H2O→Ca3Si2O7.3H2O+Ca(OH)22C2S+4H→C3S2H3+CH (in cement nomenclature)
Tricalcium aluminate (C3A) hydrates very quickly, within minutes to hours, generating large amounts of heat, to form C2AH8 and C4AH13, which then convert with time to stable C3AH6:
2Ca3Al2O6+21H2O→Ca2[Al(OH)5]2.3H2O+2[Ca2Al(OH)7.3H2O]→2{Ca3[Al(OH)6]2}+9H2O(3)2C3A+21H→→C2AH8+C4AH13→2C3AH6+9H (in cement nomenclature)
A rapid reaction immediately follows between the calcium sulphate in solution and the calcium aluminate hydrate, lasting from several minutes to hours, releasing large amounts of heat and forming ettringite:
2[Ca2Al(OH)7.3H2O]+3CaSO4.2H2O+14H2O→Ca6[Al(OH)6]2(SO4)3.26H2O+Ca(OH)2(4) C4AH13+3CSH2+14H→C3A.3CS.H32+CH
Another, fourth, component of cement is calcium aluminoferrite (C4AF), and its hydration is very similar to that of C3A. Mortar is prepared by adding sand to the cement and water mix, according to known preparation methods. Concrete is prepared by adding sand and aggregates (gravel) to the cement mix with water as well as different chemical additives and different chemical admixtures according to the desired properties. The last two ingredients, C3A and C4AF undergo an immediate reaction in the first minutes after adding the water together with the plaster. This reaction significantly affects the properties and survivability of the chemical additives.
Cement clinker is a solid material produced to production Portland cement as an intermediary product. Clinker occurs as lumps or nodules, usually 3 to 25 millimetres in diameter. It is produced by sintering (fusing together without melting to the point of liquefaction) limestone and aluminosilicate materials such as clay during the cement kiln stage.
Concrete is a fascinating material with a complex structure. At the macroscopic level, it appears as a simple two-phase composite: aggregate particles embedded within a cement paste matrix. However, zooming in reveals a third, crucial phase: the interfacial transition zone (ITZ) between the aggregate and the hardened cement paste (HCP). This microscopic realm, particularly the intricate pore system within the HCP, has become a focal point of concrete research in recent decades.
The pore system, characterized by its porosity and pore size distribution, significantly influences the strength and durability of concrete. Various methods exist to analyse these characteristics, including fluid displacement, helium psychometry, capillary condensation, adsorption-desorption isotherms, small-angle X-ray scattering (SAXS), scanning electron microscopy (SEM), nuclear magnetic resonance (NMR), AC impedance spectroscopy, mercury intrusion porosimeter (MIP), and backscattered electron imaging (BSE).
Pores come in various types and shapes, each playing a specific role. Gel pores, capillary pores, compaction pores, and ITZ pores all contribute to the overall pore volume, affecting properties like shear rate, strength, workability, and consistency of the concrete mix. Durability, however, is primarily governed by the interconnectedness of these pores. Larger pores exert a greater influence on strength and durability than gel pores, which mainly affect shrinkage and creep. Understanding the porosity and pore size distribution provides valuable insights into the overall performance of concrete. These characteristics are influenced by factors such as the water-to-cement ratio, age, and size of the cement particles.
Beyond porosity, other parameters also impact the workability and consistency of fresh concrete. These include the flow behaviour within the mixer, segregation and bleeding tendencies, homogeneity, continuity, fluidity, colour, air contact, degree of hydration, and heat of hydration rate. Continuous monitoring of these parameters is essential during concrete production.
Today, fresh concrete is typically produced in stationary concrete plants. These plants create concrete mixes with a specific slump (or flow) level and consistency, which are essential for workability. The concrete is then transported to construction sites in mixer trucks. However, several chemical and physical processes can occur during production, transportation, and even while waiting to discharge the concrete. These processes can negatively affect the concrete, primarily by reducing its slump and consistency. Several factors contribute to this, including:
These undesirable processes alter the properties of the fresh concrete mix, impacting its performance, hardening ability, and other key characteristics. To counteract these effects, it's often necessary to add water at the construction site, along with higher doses of chemical admixtures like retarders and water reducers. In some cases, additional cement or admixtures might be added as a safety measure. However, these adjustments can lead to further issues, such as producing unstable concrete mixes and compromising the final concrete quality.
The use of such subpar concrete mixes, which deviate from the desired specifications, can have detrimental effects on construction projects. Uncontrolled addition of water further aggravates the problem, adversely affecting both the performance characteristics of fresh and hardened concrete and the instability of fresh and hardened concrete. To prevent this, quality control is essential. This involves ensuring the concrete mix consistently meets the required specifications and that the final product possesses the desired properties, including strength, durability, workability, and appearance. Achieving this requires thorough testing and monitoring at various stages of the production and construction process. The quality of concrete is paramount in any construction project, as it directly influences the structure's durability, load-bearing capacity, and resistance to environmental factors.
To address these challenges, volumetric stationary concrete mixers and volumetric concrete production trucks have been introduced. These systems offer several advantages over traditional ready-mix concrete plants and trucks. A volumetric stationary concrete mixer is a highly mechanised and automated piece of equipment designed for precise concrete production. Fixed in a permanent location, it accurately weighs and mixes cement, aggregates, and water according to a specified ratio, ensuring consistent concrete quality. In contrast, a ready-mix concrete mixer involves a more complex setup. It utilises a system of storage, feeding, batching, mixing, and control devices to combine various aggregates, binders, admixtures, additives, and water in precise proportions. This centralised mixing process supplies large quantities of fresh concrete. A central control room, equipped with a computerised control system, manages and monitors the entire production process, from handling raw materials to discharging the finished concrete mix.
A ready-mix truck is equipped with a large rotating drum that transports the pre-mixed concrete from the stationary plant to the construction site. The concrete is typically batched and mixed at the plant and continues to mix during transport. Conversely, a volumetric concrete production truck allows for continuous production of concrete even during transport. This on-site mixing capability enables adjustments to the concrete mix based on factors like delivery time, environmental conditions, changes in raw material quality, and any chemical reactions or physical changes that occur during transit. Unlike a ready-mix truck, a volumetric truck has separate compartments for various chemical admixtures and water. After calibration, the truck's onboard computer can calculate the precise amount of each ingredient needed to produce any type of concrete mix with the desired properties, flow, strength, and setting time, ensuring consistent quality despite the challenges of transportation.
Essentially, large volume stationary concrete mixers and production trucks provide the advantage of producing and delivering fresh concrete on demand, as well as maintaining concrete stability throughout the entire time from production to unloading of concrete at the construction site. This eliminates the issue of the concrete becoming “hot” or prematurely setting in the drum during transport, which can necessitate the addition of water and lead to uncontrolled changes in the concrete's properties. However, even with the advantages of volumetric mixers, certain challenges persist. The concrete delivered by these trucks often arrives with either too much or too little water, resulting in a mix that is too “wet” or too “dry”. This inconsistency stems from inaccuracies in the batching system, variations in the raw materials or admixtures, inconsistencies in the timing of admixtures addition, or fluctuations in environmental conditions during production and transport. Consequently, the delivered concrete may not meet the required specifications, necessitating adjustments at the construction site. potentially compromising the concrete's quality, and adversely affecting its intended performance.
Thus, in view of the above, the long-standing problems in the concrete industry that present invention has the potential to solve:
Thus, by optimising admixture usage and minimizing cement content (through improved quality control), the invention could contribute to reducing the carbon footprint of concrete production and reducing the environmental impact, in general. Moreover, consistent concrete quality and reduced rework can improve construction productivity and project timelines. Also, the system's continuous data collection can provide valuable data-driven insights into concrete behaviour, enabling further optimization of mix designs and production processes.
Overall, the present invention addresses several critical challenges in the concrete industry. By enabling autonomous, proactive AI-driven quality control during transportation, it has the potential to revolutionise concrete production, leading to improved quality, cost savings, and environmental benefits.
WO 2022/249162 A1, which is the publication of the co-pending application by the same inventors, describes controlling, image analysis and continuous visual monitoring of aggregates and sand processing, and physical properties and workability of fresh concrete and concrete mixes, which are manufactured from the aggregates and sand, and then transported to construction sites and used there for construction purposes. The system of WO 2022/249162 A1 comprises visual monitoring devices, stationary or mobile, installed or remotely used at quarries, concrete plants, in concrete trucks and at the construction sites. The method for continuous visual monitoring of the aggregates, sand and concrete described in WO 2022/249162 A1 is based on image or video processing and analysis of the aggregates and sand, fresh concrete, concrete mixes or precast concrete, the concrete slump levels, segregation and bleeding, homogeneity of the mixture and consistency.
CN 114295732 A focuses on segregation only and solves the problem that the segregation degree of the concrete exceeds the standard in the transportation process of the mixer truck. It describes the use of an acoustic sensor to monitor the segregation degree of a concrete mixer, and uses a pre-trained segregation degree model, where matching the adaptive rotating speed corresponds to the segregation degree through a preset rotating speed and segregation degree adaptive parameter setting library. Based on this, it is capable of adjusting the rotating speed of the mixing drum of the mixer according to the adaptive rotating speed.
WO 2009/126138 A1 describes a method for monitoring thixotropy in concrete mixing drum by measuring the reversible, time-dependent reduction in viscosity occurring when concrete is subjected to mixing, and employs a mixing drum and conventional slump monitoring equipment as used on ready-mix trucks. WO 2009/126138 A1 suggests to “train” the control system based on historical batch data of the relationships and correlations between a particular concrete mix design, the effect of particular additions (water, chemical admixtures) on the particular concrete mix and batch proportions.
US 2022/355509 A1 describes a system and method for monitoring fresh concrete being handled in a concrete mixer using trained data processing engines. This publication uses a reactive approach, focuses on reactive monitoring of fresh concrete properties and detecting abnormal operating conditions. It uses sensors and a “trained data processing engine” to determine properties like slump and output signals or alerts based on the measurements. In addition, US 2022/355509 A1 implies a system that uses basic statistical analysis or machine learning to associate sensor data with concrete properties. Also, US 2022/355509 A1 is silent about minimising water addition to maintain concrete quality. It appears to treat water as just another variable to be monitored. Although US 2022/355509 A1 mentions various sensors (rheological probe, drum speed sensor, temperature sensor), it does not specify how the data from these sensors are combined and used for decision-making.
The present invention solves long-standing problems in the concrete industry, such as variable quality during production and transport, and the uncontrolled addition of water at the construction site. By employing a sophisticated, proactive AI system, this invention automates concrete production, prioritising minimal water usage and consistent quality throughout the entire process, from initial mixing to final delivery.
The present invention relates to a concrete production system comprising:
In a further embodiment, the AI-based control system is configured to determine a quantity of water to add to the concrete based on the real-time sensor data received from the continuous monitoring system.
In an additional embodiment, a method for producing concrete, comprising:
The sensors used in a continuous monitoring system of the invention are selected from:
Thus, the present invention presents a valuable solution in the concrete production industry. By addressing the challenges of maintaining concrete quality during production and transportation and eliminating the need for uncontrolled water addition at the construction site, the system of the present invention offers significant benefits in terms of cost savings, improved concrete quality, and reduced environmental impact.
The key inventive aspects of the present invention are:
In one embodiment of the present invention, the concrete production platform is a stationary platform operated by an external operator or an autonomous operating system. In another embodiment, the concrete production platform is a mobile platform operated by a driver, an external operator, or an autonomous operating system for transporting components of the system. In some embodiments, the dispensing mechanism comprises dispensers, flow meters, and nozzles for controlled and continuous measuring, dosing, and dispensing of the admixtures and water into the mixer tank as directed by the AI-based control system.
Various embodiments may allow various benefits and may be used in conjunction with various applications. The details of one or more embodiments are set forth in the accompanying figures and the description below. Other features, objects and advantages of the described techniques will be apparent from the description and drawings and from the claims.
The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.
Disclosed embodiments will be understood and appreciated more fully from the following detailed description taken in conjunction with the appended figures. The drawings included and described herein are schematic and are not limiting the scope of the disclosure. It is also noted that in the drawings, the size of some elements may be exaggerated and, therefore, not drawn to scale for illustrative purposes. The dimensions and the relative dimensions do not necessarily correspond to actual reductions to the practice of the disclosure.
In the following description, various aspects of the present application will be described. For purposes of explanation, specific details are set forth to provide a thorough understanding of the present application. However, it will also be apparent to one skilled in the art that the present application may be practiced without the specific details presented herein. Furthermore, well-known features may be omitted or simplified in order not to obscure the present application.
The term “comprising”, used in the claims, is “open ended” and means the elements recited, or their equivalent in structure or function, plus any other element or elements which are not recited. It should not be interpreted as being restricted to the means listed thereafter; it does not exclude other elements or steps. It needs to be interpreted as specifying the presence of the stated features, integers, steps, or components as referred to, but does not preclude the presence or addition of one or more other features, integers, steps or components, or groups thereof. Thus, the scope of the expression “a composition comprising x and z” should not be limited to compositions consisting only of components x and z. Also, the scope of the expression “a method comprising the steps x and z” should not be limited to methods consisting only of these steps.
Unless specifically stated, as used herein, the term “about” is understood as within a range of normal tolerance in the art, for example, within two standard deviations of the mean. In one embodiment, the term “about” means within 10% of the reported numerical value of the number with which it is being used, preferably within 5% of the reported numerical value. For example, the term “about” can be immediately understood as within 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.05%, or 0.01% of the stated value. In other embodiments, the term “about” can mean a higher tolerance of variation depending on for instance the experimental technique used. Said variations of a specified value are understood by the skilled person and are within the context of the present invention. As an illustration, a numerical range of “about 1 to about 5” should be interpreted to include not only the explicitly recited values of about 1 to about 5, but also include individual values and sub-ranges within the indicated range. Thus, included in this numerical range are individual values such as 2, 3, and 4 and sub-ranges, for example from 1-3, from 2-4, and from 3-5, as well as 1, 2, 3, 4, 5, or 6, individually. This same principle applies to ranges reciting only one numerical value as a minimum or a maximum. Unless otherwise clear from context, all numerical values provided herein are modified by the term “about”. Other similar terms, such as “substantially”, “generally”, “up to” and the like are to be construed as modifying a term or value such that it is not an absolute. Such terms will be defined by the circumstances and the terms that they modify as those terms are understood by those of skilled in the art. This includes, at very least, the degree of expected experimental error, technical error and instrumental error for a given experiment, technique or an instrument used to measure a value.
As used herein, the term “and/or” includes any and all combinations of one or more of the associated listed items. Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the specification and relevant art and should not be interpreted in an idealized or overly formal sense unless expressly so defined herein. Well-known functions or constructions may not be described in detail for brevity and/or clarity.
The present invention describes a concrete production system comprising:
Thus, the present invention centres on a concrete production system equipped with a proactive AI-powered quality control system. The invention applies to concrete production in a stationary mixer or during transportation on any suitable mobile platform. The AI system continuously monitors the state of the concrete within the mixer during production or transportation. Using proactive AI, the system autonomously determines the optimal timing, type, and quantity of chemical admixtures and water to be added, ensuring the concrete maintains desired properties without manual intervention.
The system of the present invention is capable of autonomously controlling the addition of admixtures based on real-time monitoring and AI decision-making. It continuously monitors in real time concrete properties during the concrete production and transportation, enabling timely adjustments. The system of the present invention is entirely AI-driven, based on machine learning models, and it drives the decision-making process for admixture addition.
The system of the present invention aims to maintain concrete quality by primarily adjusting admixtures and essentially minimising the addition of water. The AI model suitable for the autonomous admixture control system in the concrete production mixer is a hybrid model combining reinforcement learning (RL) and supervised learning. Its core components are:
The RL agent's goal is to learn the optimal policy for adding admixtures and water, maximising concrete quality while minimising water usage. Its state encompasses real-time sensor data (temperature, slump, etc.), concrete mix design parameters, environmental conditions, and time elapsed since mixing. The action space consists of decisions on the type, quantity, and timing of admixture/water addition. The reward function reflects the concrete's quality, penalising deviations from desired properties and excessive water usage. It incorporates factors like compressive strength, workability, setting time, and cost efficiency.
The SL model's objective is to predict concrete properties based on sensor data, admixture history, and other relevant factors. This model aids the RL agent in decision-making. Historical data from concrete production, including sensor readings, admixture additions, and resulting concrete properties is used as training data. The model type is a regression model (e.g., neural network, random forest) to predict continuous properties like slump or compressive strength, or a classification model for discrete properties like setting time categories.
The AI workflow of the present invention comprises the following stages:
The hybrid AI model of the present invention is highly adaptable. The RL agent allows the system to learn and adapt to varying conditions and concrete mix designs. It is also highly predictable. The SL model helps the RL agent anticipate concrete property changes, enabling proactive admixture adjustments. In addition, the RL is capable of optimizing admixture usage for cost-effectiveness and environmental friendliness. Also, the AI hybrid model of the present invention allows for real-time decision-making during concrete transportation. A realistic simulation environment can be valuable for initial RL agent training and testing, reducing risks in real-world deployment. Techniques like attention mechanisms or model-agnostic explanations provide insights into the AI's decision-making, increasing trust and facilitating troubleshooting. Special safeguards are implemented in the AI hybrid model of the invention to prevent the AI from making decisions that could compromise concrete quality or safety.
In some embodiments, the reinforcement learning agent uses the predictions from the supervised learning model to make the determination. In other embodiments, the AI-based control system is configured to determine a quantity of admixtures and water to add to the concrete based on the data received from the continuous monitoring system.
In a further embodiment, the proactive AI-based control system of the invention comprises:
In some embodiments, the proactive AI-based control system uses machine learning models to predict admixture addition requirements. In another embodiment, the reinforcement learning agent uses the predictions from the supervised learning model to make the autonomous determination. In a certain embodiment, the method of the present invention further comprises autonomously determining a quantity of water to add to the concrete based on the data received from the continuous monitoring system.
The embodiments of the present invention describe the real-time monitoring and control during the concrete production and transportation, whereas the above acknowledged prior art focuses on initial mix optimisation or monitoring at the batch mixer. The embodiments explicitly state the use of a continuous monitoring system with a specific combination of sensors, whereas the above acknowledged prior art mentions sensors in general but do not specify the combination or how the data is used. The proactive AI-based control system's ability to autonomously determine admixture type, quantity, and timing clearly distinguishes the present invention from systems that rely on user input or simple historical data analysis. Moreover, the present invention prioritises minimizing water addition, whereas the above acknowledged prior art is silent about water reduction during the concrete production and transportation.
By definition, proactive AI is an artificial intelligence system that anticipates and fulfils the needs of users without prompting. This type of AI uses the latest algorithms, machine learning (ML), and predictive analytics to forecast future commands and proactively respond to them. Reactive AI operates in the moment. It analyses incoming data, compares it to its knowledge base, and then reacts accordingly. It can be thought as a reflex action. It does not predict or anticipate future events; it simply responds to the current situation. In contrast, proactive AI, on the other hand, anticipates future needs and events. It leverages historical data, predictive models, and real-time information to make decisions and take action before an event occurs. This allows it to optimise processes, prevent problems, and capitalise on opportunities. Reactive AI focuses on reacting to current situations, whereas proactive AI anticipates and prevents future events. Reactive AI uses primarily current data and makes decisions based on immediate context, whereas proactive AI uses historical, real-time, and predictive data and makes decisions based on predictions and long-term goals. Thus, reactive AI only responds to events, whereas proactive AI initiates actions.
For example, a sensor controlled by reactive AI can only detect a specific parameter, compare it to a predefined threshold, and then trigger an alert for the concrete production system operator. In contrast, the proactive AI of the present invention analyses historical sensor data, creates system performance logs, considers external factors like weather and temperature, and uses this information to predict when concrete quality is likely to decline. It then proactively schedules the autonomous addition of a specific admixture to the concrete mixer, determining both the amount and timing of the addition.
In one embodiment of the present invention, the concrete production platform is a stationary platform operated by an external operator or an autonomous operating system. In another embodiment, the concrete production platform is a mobile platform operated by a driver, an external operator, or an autonomous operating system for transporting components of the system. In some embodiments, the dispensing mechanism comprises dispensers, flow meters, and nozzles for controlled and continuous measuring, dosing, and dispensing of the admixtures and water into the mixer tank as directed by the AI-based control system.
In another embodiment, said continuous monitoring system comprises at least two sensors selected from the group consisting of an imaging camera, a hydraulic pressure gauge, a temperature gauge, and an acoustic sensor.
In a further embodiment, said imaging camera, for example a thermal image camera, is installed inside the concrete mixer tank for continuously gathering visual information, thermal information, and thermal profile of the concrete at any time during production, before transportation, during transportation, prior to discharge and during the discharge of the concrete through a mixer trough at a construction site.
The acoustic senor is installed on the mixer tank for continuously examining changes in a sound level (dB), frequency (Hz) and duration, and a sound of low and full load of the concrete inside the concrete mixer, said acoustic senor is thus configured to monitor the workability, cohesion, homogeneity, segregation, and water separation of the concrete.
The hydraulic pressure gauge is installed for indicating a hydraulic pressure of the concrete inside the concrete mixer tank and a hydraulic load intensity on the mixer motor during loading and prior to discharge of the concrete while mixing at a high rotation frequency of the mixer tank from about 5 rpm to about 95 rpm, and during transportation while mixing at a low rotation frequency from about 1 rpm to about 4 rpm. The hydraulic pressure and hydraulic load intensity are indicators of the workability of the prepared concrete, and said hydraulic pressure gauge is thus configured to provide an indication to simulate the workability of the concrete.
The temperature gauge is installed for continuously monitoring and controlling the concrete temperature and surrounding temperature outside the mixed concrete, said at least one temperature gauge is thus configured to monitor a hydration progress, including the degree of hydration, rate of heat of hydration and slump or flow reduction of the concrete, and water absorption by aggregates of the concrete.
After preloading aggregates, including sand, gravel, and crushed stone rock, to the stationary mixer or the truck mixer tank in the concrete plant, chemical additives, for example, fly ash or slag, water, and a chemical clinker (including gypsum addition) used as a binder (cement) for producing concrete upon mixing with water, the required chemical admixtures are disposed in the corresponding reservoirs on the concrete production platform.
The chemical admixtures are proactively added to the mixer at a certain rate, at different times during the production and transportation process, and the dosages are controlled by the AI-based system developed in accordance with the present invention, depending on the properties of the raw materials in the concrete, the progress of the chemical reaction of the cement with water, and chemical and physical changes that occur during the transport of the concrete to and from the construction site.
The autonomously controlled and continuous addition of the particular types of the chemical admixtures, in specific dosages, and at particular intervals of time is one of the major aspects of the present invention. In fact, such addition of the chemical admixtures in the proactively controlled and continuous manner obviates the use of water in the production of the concrete in stationary mixers or on the go and make the entire concrete production process much more efficient and allows the full automation of the production process.
Furthermore, the mobile platform does not need to transport large volumes of water, in contrast to a conventional concrete mixing truck. Carefully controlled and continuous addition of the chemical admixtures without water makes it possible to prepare the fresh concrete or batched concrete mix and maintain the required and desired physicochemical properties of the concrete, its stability, quality, and homogeneity during the transportation and then during the discharge of the prepared concrete at the construction site. In the present invention, only small amounts of water are added from a small water container installed on the mobile platform to wash the residuals of the dispensed dosage of a chemical admixture into the mixing tank, thereby increasing accuracy of the dosing and dispensing of the chemical admixtures.
Thus, the concrete production process of the present invention is fully controlled and adjusted by an “autonomous operating system”, which is a proactive AI-based control system of the present invention that enables the fully autonomous operations for an unmanned stationary or mobile platform and continuous monitoring and quality control. The AI-based system of the invention make the decisions in accordance with the properties of the raw materials, such as aggregate water absorption rate, aggregate moisture, quality of the aggregates and sand, presence of impurities, such as dust or clay, in the raw materials, hydration rate, transportation time, hydration progress of the different cement component and fineness of the cement, external temperature and humidity conditions, and the desired physicochemical properties of the obtained concrete.
The following physicochemical parameters of the produced concrete and of the concrete production process are continuously monitored by the concrete-monitoring and quality-control system, and adjusted, if needed:
In general, the rate of heat of hydration of cement in the concrete indicates the viscosity of the concrete and determines an amount of a hydration stabiliser to be added to the cement in the mixer tank. The hydration stabiliser is one of the chemical admixtures formulated to retard the concrete production over extended periods of time or on the other hand, to add accelerator admixture to decrease the setting time and also to prevent the freeze of water in cold areas to achieve faster setting and increased an initial strength. The heat of hydration of cement is heat evolution, which is proportional to the change in viscosity during the concrete production process.
In certain embodiments of the present invention, the above physicochemical parameters of the produced concrete are correlated with an amount of water to add to the concrete in the concrete mixer tank in order to reach a required water-to-cement ratio and not to exceed this ratio. The physicochemical parameters of the produced concrete are also correlated with an amount and types of the different chemical admixtures to add to the produced concrete at predetermined dosages and intervals of times, to disperse said concrete and thereby increase the slump level of the concrete to the desired slump level, without adding water.
Reference is now made to Table 1 that provides a framework for the AI model's operation of the present invention. The implementation of this framework involves complex algorithms and decision-making processes. However, this illustrates the core aspect of using combined sensor data and proactive AI to autonomously control admixture addition in real-time during the concrete production and transportation.
In this table, while the primary focus is on real-time, non-destructive sensing, the slump test and ait content test can be incorporated periodically for calibration or validation. Specific admixture dosages are determined by the AI model based on the magnitude of the parameter change, concrete mix design, and other relevant factors. The proactive AI model continuously learns and refines its decision-making based on the outcomes of its actions and any additional data collected. In addition, the proactive AI system implements safeguards to prevent the AI from taking actions that could compromise concrete quality or safety. Human oversight or intervention may be necessary in certain situations.
In one embodiment of the present invention, the AI input sensor data comprises images or video frames of the concrete in the mixer tank from an imaging camera; real-time hydraulic pressure readings from a hydraulic pressure gauge; concrete and ambient temperature measurements from a temperature gauge; sound level, frequency, and duration data from an acoustic sensor; optionally an aggregate moisture content at loading from a moisture sensor; and optionally a mixer tank rotation speed from and RPM gauge. In another embodiment of the present invention, the AI input contextual data comprises target slump, strength and setting time from (concrete mix design); aggregate properties including type, size distribution and moisture content; cement type including hydration characteristics, environmental conditions, such as temperature and humidity; time elapsed since initial mixing; and admixture history including admixture types and quantities already added.
A historical data fed to the AI system may be the following parameters selected from:
In a further embodiment of the present invention, the AI output comprises decisions on type of admixture to add (if any), quantity of admixture to add, timing of admixture addition, and quantity of water to add (if any). In addition, the AI of the invention may also output the levels of and deviations from the desired quality and stability of the produced concrete in the concrete mixer tank during the transportation and prior to the discharge, are characterised by one or more parameters:
In yet further embodiment of the present invention, the proactive AI comprises at least one of the following deep layers:
In the present invention, these layers work together as follows. Sensor data processing is realised through CNNs processing images and RNNs/Transformers handling time-series sensor data. Feature extraction and integration is performed with MLPs combining and transforming features from different sensors and models. The RL network uses the integrated features and contextual data to make admixture/water addition decisions. The outcomes of the AI's actions (concrete quality, resource usage) are fed back as rewards to the RL agent, enabling it to learn and improve its policy over time (feedback loop).
Pre-training CNNs on large image datasets and RNNs/Transformers on relevant time-series data significantly improves the overall performance. If data is limited, transfer learning from related domains (e.g., material science) can help initialise the models. In order to understand the AI's decision-making process, techniques like attention mechanisms or model-agnostic explanations can be incorporated. Safeguards are implemented to prevent unsafe or undesirable actions. In addition, the AI model of the present invention is robust to noisy or incomplete sensor data.
The monitoring and adjusting process is continuous, which means it is automatically carried out by the AI-based system of the present invention until the concrete is discharged (offloaded) or prior to that, if so desired. Reference is now made to
In the present application, the term “volumetric concrete production” means the production of concrete, which is mixed and delivered to the construction site by volume of concrete, rather than weight. Volumetric concrete in the present invention is autonomously produced from various ingredients (water, cement, additives as fly ash, slag, limestone powders etc., sand, and aggregates) in self-contained portable batch mixers, which produce concrete by proportioning the materials out over time by volume and relating that volume back to the materials specific weight. Volumetric concrete production offers complete control over when, where, how much, and what type of concrete is mixed and applied for any type of project, large or small. That flexibility to adapt to any situation is unmatched by any other approach.
The term “fresh concrete” means that the concrete had been recently mixed from the beginning of loading the concrete in the plant, transporting the concrete, discharging the concrete, and completing the application of the concrete on the concrete element. It has the required homogeneity and consistency, and it possess its original workability at any time and state before transportation, during transportation, prior to discharge and during the discharge of the concrete through a mixer trough at a construction site, so that it can be placed, handled, consolidated, and finished by the intended methods. Concrete is referred to as “fresh” when the setting and hardening process has not yet started. “Fresh concrete” can be deformed and poured which means it can be transported or pumped and used to fill moulds or formwork. It appears in plastic state and can be moulded in any forms, whereas the hardened concrete is the one which is fully cured. For the concrete to be considered “fresh”, it should be easily mixed and transported, be uniform throughout a given batch and between batches, and be of a consistency so that it can fill completely the forms for which it was designed.
“Batched concrete mix” means that the concrete was mixed from the required concrete ingredients with either weight or volume according to the mix requirement of a consistent quality of concrete. To produce the batched concrete mix, the ingredients should be loaded into the mixer in a predefined sequence and amount. Two main types of batch mixers can be distinguished by the orientation of the axis of rotation: horizontal or inclined (drum mixers) or vertical (pan mixers). The drum mixers have a drum, with fixed blades, rotating around its axis, while the pan mixers may have either the blades or the pan rotating around the axis. In the present invention, both types of mixers can be used to produce the batched concrete mix.
In general, fresh concrete and batched concrete mix are autonomously produced in the present invention from a combination of aggregates, including sand, gravel, and crushed stone rock of different sizes, water, a chemical clinker (including gypsum addition) used as a binder for producing cement upon mixing with water, chemical additives, for example fly ash, limestone powder or slag, and chemical admixtures. The main properties of concrete are:
Unless otherwise defined, “homogeneity” of a fresh concrete or a batched concrete mix is a percentage according to the given composition of components. The concrete mix is considered homogeneous if the samples taken from different places in the mixer contain the individual components of the mixture in equal percentages. The concrete mix homogeneity is associated with the strength of concrete and assessed by engineers using visual inspection and experience. “Concrete consistency” in the present invention refers to the relative mobility or ability of freshly mixed concrete to flow. It includes the entire range of fluidity from the driest to the wettest possible mixtures. Plastic consistency indicates a condition where applied stress will result in continuous deformation without rupture. “Slump”, “slump level”, “flow” or “flow level” is the measure of concrete homogeneity, consistency, and fluidity during the condition of the fresh concrete. It shows the flow and overall workability of freshly mixed concrete.
“Concrete workability” is a term that refers to how easily freshly mixed concrete can be placed, consolidated, and finished to a homogeneous condition with minimal loss of homogeneity. In general, the workability of concrete is determined by how fluid the concrete mix is (i.e., as the cement-to-water ratio), which is essentially the slump of concrete. It is synonymous with placing ability and involves not only the concept of a consistency of concrete, but also the condition under which it is to be placed, i.e., size and shape of the member, spacing of reinforcing, or other details interfering with the ready filling of the forms.
The more fluid the concrete, the higher the slump, and whilst the slump is seen as a measure of water content, it is typically also used as a measure of concrete consistency. Simply put, the higher the slump, the wetter the mix. Five-inch slump is very common with normal weight concrete and is a good for pumping. Slumps that are above average will cause reduced strength, durability, and permeability of the concrete, if more water is added to increase the slump level.
There are three primary factors that affect the workability of concrete:
As used herein, the term “chemical admixture” includes chemical adjuvants added during continuous and ongoing concrete mixing to enhance or to adjust the workability (slump) of the fresh concrete or to affect other physicochemical properties of the concrete as mentioned above (setting times, homogeneity). Chemical admixtures are added to concrete batch during mixing concrete according to the progress of the hydration, aggregates water absorption, environmental conditions, and chemical and physical properties of the raw materials in concrete, over the time and transportation of the concrete to the construction sites and at the sites before unloading the concrete. They improve concrete quality, adjust the required workability, manageability, acceleration, or retardation of setting time, among other properties that could be altered to get specific results. Non-limiting examples of the dispersants suitable for use in the present invention are polycarboxylate polymer and naphthalene sulphonate.
In particular embodiments, the chemical admixtures used in the present invention are selected from the group consisting of:
In particular regard, a “cement accelerator” is an admixture for the use in concrete, mortar, plasters, or screeds. The addition of an accelerator speeds the setting times (initial and final) and thus, cure time starts earlier. This allows concrete, for example, to be placed in winter with reduced risk of frost damage and in a shorter time. Concrete is damaged if it does not reach to the required setting times or to a strength of 10-25 MPa before freezing. Typical chemicals regularly used for acceleration are calcium nitrate (Ca(NO3)2), sodium thiocyante, sodium silicate, calcium nitrite (Ca(NO2)2), calcium formate (Ca(HCOO)2) and aluminium compounds. Novel alternatives include cement based upon calcium sulphoaluminate (CSA), which sets within 20 minutes and develops sufficient rapid strength that an airport runway can be repaired in a six-hour window and be able to withstand aircraft use at the end of that time, as well as in tunnels and underground, where water and time limitations require extremely fast strength and setting.
The slump test is one of the tests used to measure the workability and assess the consistency of fresh concrete. There are other techniques to test workability such as a flow for very high workability concrete, such as self-compacting concrete type (SCC). Generally, it is used to check that the correct volume of water has been added to the mix. Conventionally, workability of concrete is determined by checking the slump level of concrete using a cone as shown in
In the present invention, the slump test is carried out directly on the concrete production platform of the invention. Slump test results can be classified in four types:
Unless otherwise defined, “concrete stability” in the present invention refers to the ability of the produced concrete to remain stable and homogeneous during handling, transportation, and discharge at the constructive sites without excessive segregation. Stability of the concrete is characterised by the bleeding and segregation tendencies of the concrete using a direct method of measurement, for example the method based on floatation over carbon tetrachloride.
Customers and buyers of concrete usually order a certain mechanical strength and slump level of the concrete according to the required specifications of the concrete to ensure its quality. Concrete systems produce concrete and supply it to the customers. The slump level of concrete usually decreases from the moment the concrete is prepared due to the hydration of cement in the concrete with water, water absorption by aggregates, admixtures absorption by aggregates in the concrete, change of ambient temperature, impurities such as clays and organic materials in the aggregates and sand used to make the concrete, reduction performances of the different types of admixtures mixed during the load of the cement water and aggregates in the initial mixing in the concrete system. Therefore, the slump level of concrete indicating its homogeneity, consistency and fluidity decreases with the time of production and transportation of the concrete until the concrete arrives at the construction site and is used.
Right after the mixing of all the concrete components, or during production and transportation of the concrete with a certain slump level in the concrete-production system of the present invention and during the discharge of the concrete at the construction site, it is possible, using the continuous monitoring system of the invention, comprising at least two sensors selected from an imaging camera, a hydraulic pressure gauge, a temperature gauge, and an acoustic sensor, to determine or to simulate the concrete slump and the concrete slump reduction of the produced concrete.
The slump level and other characteristics of the freshly prepared concrete can be assessed by AI-controlled sensors as the image processing and finding a correlation between the image of the concrete fluidity and the slump test performed as mentioned above and further detailed. Reference is now made to
Apart from the slump level of concrete, the quality of concrete is affected by a number of factors.
In a certain embodiment of the present invention, a continuous monitoring system of the present invention configured to monitor, during production of the concrete, one or more properties of the concrete within the concrete mixer tank, comprises at least two sensors selected from the group consisting of:
An exemplary thermal imaging camera that produces images, videos, thermograms and thermal profiles of the concrete, used in the system of the present invention is a forward-looking infrared (FLIR) camera, which is capable of monitoring the consistency of the fresh concrete or concrete batched mix. This type of cameras does not “see” water in the concrete, but rather visualises the impact water has on the temperature of surfaces around them due to the evaporation process.
In another embodiment, the continuous monitoring system of the present invention further comprises a tachometer or a revolutions-per-minute (RPM) gauge installed on the mixer for indicating a centrifugal force or rotation speed and tracking progress of the concrete mixer tank, and additional simulation of the slump level.
In the present application, the terms “tachometer” and “RPM gauge” are considered entirely equivalent and used therefore interchangeably. In general, the RPM gauge or tachometer is a device measuring the centrifugal force or rotational speed of a shaft or disk, as in a motor or other machine. In the concrete mixer truck, the RPM gauge measures the centrifugal force or rotational speed of the concrete mixer tank of the truck. This device usually displays the revolutions per minute (RPM) on a calibrated analogue dial, but digital displays are increasingly common and also can be used to indicate mixing or unloading of the concrete and to evaluate the volume remaining in the mixing tank.
The hydraulic pressure gauge and RPM gauge installed on the concrete mixer allow an additional indication to simulate the slump level. The control system of the present invention determines the slump level and the slump reduction and an amount of the concrete admixture, which should be added in order to increase and adjust the slump level to the desired level without adding water to the concrete. In addition, the control system determines the volume of concrete left in the concrete tank by calculating the estimated volume discharged (offloaded) by the number of the discharge rounds and a number of empty blade spiral revolutions.
In yet further embodiment, the system of the present invention further comprises a communication module installed into or connected to the computing unit and configured to:
In some embodiments, the communication module is a wireless connection module. It can be either Bluetooth® or NFC providing the short-range wireless communication between the computing unit and an external storage device or the user's interface for up to 20 m. If this module is Wi-Fi, the connection can be established with a network for up to 200 nm, while GSM allows the worldwide communication to a cloud. The external storage device or user's interface may be any mobile device or gadget, such as a smartphone or smart watch. It may also be a desktop computer, server, remote storage, internet storage or cloud. The communication module may be a wireless connection module used as a standalone device or integrated in the computing unit or in the external storage device.
Thus, the present application describes the AI-driven control and monitoring system for constantly examining the concrete slump and homogeneity, monitoring the decrease of fluidity of the concrete as a function of time and correlating it to the slump or flow level ordered by the contractor. This system proactively monitors the physicochemical properties of fresh concrete by assessing its slump level and homogeneity using image processing and identifying the reasons of the concrete failure while it is still being produced in the concrete system, transporting the concrete in the concrete truck to the construction site and at the time of discharging the concrete into the building structure or pump. The monitoring and adjusting process is autonomous and continuous, which means it is autonomously and continuously carried out by the AI-driven system of the invention until the concrete is discharged (offloaded) or prior to that, if so desired.
The AI-driven control and monitoring system allows having a regular image, a video or a thermogram of the concrete to be obtained at any given time and makes it possible, by processing the image, to autonomously assess the slump or flow level of the concrete at any given moment and without checking by an operator, truck driver or quality controller at the construction site, or to proactively detect conditions of defective concrete preparation and improper handling of the concrete mixes containing non-homogeneous concrete, lumps, water bleeding, segregation of aggregates, the slump level too high or too low, and the like.
In some embodiments, the data generated by the AI-driven control and monitoring system is dependent on thermogram parameters, time, and time intervals of the audible (acoustic) signals, temperature and temperature gradient, and hydraulic pressure.
In a further embodiment of the present invention, the concrete monitoring and quality control system is installed together with other components of the production system on the stationary platform or mobile platform of a truck suitable for transporting the fresh concrete or the batched concrete mix to a construction site and automatically discharging (offloading) the concrete at the construction site.
In yet further embodiment, the concrete monitoring and quality control system of the present invention further comprises an imaging or video camera installed on the mobile platform outside the concrete mixer tank, for monitoring events and activities outside the mixer truck. These events and activities outside the mixer truck comprise activities of factory and construction personnel, factory and laboratory workers and engineers taking samples of the discharged concrete for determining the quality of the concrete, and an operator and driver of the mobile platform.
The adjustment of the slump or flow level is enabled by four major steps of a method encoded by an algorithm of the present invention:
In most of the embodiments, the artificial intelligence involves a training process that includes training the machine-learning model with the aforementioned input data sets of the present invention received from the sensors, wherein each data set is based on a single time stamp and represents the predictions that will be made by the trained machine-learning model. This training of the machine-learning model correlates the input data with pre-determined labels, including the required quality, consistency, and stability of the fresh concrete or the batched concrete mix being produced in the mixer tank during the transportation and prior to the discharge; decrease in quality of the aggregates and change in composition of the produced concrete; a computed volume of the concrete in the concrete mixer tank; a concrete temperature; sound (audible) parameters changes that indicate “drying” and homogeneity of the concrete; required physicochemical parameters of the produced concrete and deviations from the physicochemical parameters of the concrete production process. After being trained, the proactive machine-learning model (e.g., a deep-neural network) predicts a set of actions including adding certain chemical admixtures at specific amounts and at particular intervals of time into the mixer tank to maintain the desired quality and stability of the produced concrete in the concrete mixer tank.
The levels of and deviations from the required physicochemical parameters of the produced concrete and the corresponding required examples of actions are summarised in the following table:
Using a closed control circuit that receives slump data and slump decrease at any time, it is possible to adjust the concrete mixture fluidity and homogeneity by adding a suitable chemical admixture, such as a chemical dispersant, to ensure concrete supply at the desired slump level and without uncontrolled addition of water at the construction site. This will impart much better control of the quality of the concrete.
In another aspect of the present invention, a method for producing concrete, comprising:
In the method of the present invention, the proactive AI-based control system comprises a reinforcement learning agent configured to learn an optimal policy for adding the admixture based on the data received from the continuous monitoring system; and a supervised learning model configured to predict temporal changes in the one or more properties of the concrete during the transporting based on historical concrete production data. In some embodiments, the reinforcement learning agent uses the predictions from the supervised learning model to make the determination. In other embodiments, the method of the present invention further comprises autonomously determining a quantity of water to add to the concrete based on the data received from the continuous monitoring system.
In a specific embodiment, the concrete physicochemical parameters in the method of the present invention are selected from the group consisting of:
In a certain embodiment, the concrete physicochemical parameters in the method of the present invention are correlated in the AI system:
In some embodiments, the proactive AI system used in the method of the present invention involves a training process that includes training the machine-learning model with the aforesaid input data sets, each data set is based on a single time stamp and represents the predictions that will be made by the trained machine-learning model. Said training of the machine-learning model correlates the input data with pre-determined labels, including the quality, consistency, and stability of the concrete being produced in the mixer tank during the transportation and prior to the discharge; decrease in quality of the aggregates and change in composition of the produced concrete; a volume of the concrete in the concrete mixer tank; a concrete temperature; sound changes that indicate drying of the concrete; and deviations from physicochemical parameters of the concrete production process.
In a specific embodiment, the exemplary concrete production process of the present invention, including the sensing, control, and operation of the system comprises the following actions:
To sum up, the system and the method of the present invention allow:
The concrete monitoring and quality control system of the present invention has a number of notable pros:
In a specific embodiment, the type of concrete produced by the method of the present invention is selected from the group consisting of:
A hot summer day and a cool night for concrete deliveries were selected. Identical concrete mix designs and trucks for both the proactive AI-controlled system and the traditional system were used. Both trucks were equipped with temperature sensors to continuously monitor the concrete temperature during transit. In the AI-controlled truck, the system adjusts admixtures (retarder in hot weather, accelerator if needed in cold) based on real-time temperature data to maintain the target slump. In the traditional truck, no adjustments are made during transport. Upon arrival at the construction site, the slump of concrete from both trucks were measured, and the amount of water added at the site to achieve the target slump in each case was recorded.
The reference is now made to
A t-test was used in statistical analysis to compare the average water added in each condition (hot/cold) between the two systems. The t-test results (p<0.01 for hot day, p<0.05 for cool night) indicate that the reduction in water addition achieved by the AI-controlled system is statistically significant in both temperature conditions. This means the difference is likely not due to random chance and reflects a true advantage of the invention. None of the prior art references teach or suggest a system capable of actively counteracting temperature effects on concrete properties during transportation. The statistically significant reduction in water addition achieved by the present invention demonstrates a non-obvious ability to maintain consistent slump without compromising concrete quality, a capability not found in the prior art.
Indeed, prior art focuses on initial mix optimisation, but this experiment clearly demonstrates the AI system's ability to actively counteract temperature effects during transit, a capability not suggested in the prior art. The significant reduction in water addition highlights the surprising effectiveness of the invention in preserving concrete quality, leading to higher strength, improved durability, and reduced material waste.
Two concrete trucks were intentionally delayed (one with AI control, one traditional) for a significant period (2 hours). Concrete properties were monitored (slump, setting time) in both trucks using appropriate sensors. In the AI-controlled truck, the system detects early setting and dispenses retarders to extend workability. In the traditional truck, no adjustments are made. Upon arrival at the construction site, the workability of the concrete from both trucks was assessed.
The experiment demonstrated that the proactive AI-controlled concrete maintains workability even after the delay, allowing for successful placement. In contrast, the traditional concrete showed significant loss of workability or even becomes unplaceable due to premature setting. Table 4 below summarises these obtained results.
While not quantifiable with a t-test, as the outcome is more binary (placeable vs. unplaceable), the 100% success rate of the proactive AI-controlled system in maintaining placeable concrete after a 2-hour delay is a striking result. This stark contrast to the 0% success rate of the traditional method strongly suggests a non-random, significant advantage. Indeed, the difference clearly and potentially repeats the experiment to establish a trend.
Prior art does not address the challenge of unexpected delays during transportation, but focuses on adjusting mixer speed to address segregation. This experiment demonstrates the proactive AI system's ability to prevent costly waste by actively responding to early setting, a capability not suggested in the prior art. The ability of the present invention to prevent premature setting and maintain workability in such situations is a non-obvious solution not hinted at in the prior art. This translates to significant cost savings by reducing material waste, disposal costs, and project delays.
Two batches of concrete with the same mix design but using aggregates with significantly different moisture contents (one dry, one pre-wetted) were prepared. The concrete was delivered using both the proactive AI-controlled and traditional systems. Concrete properties (slump, air content) were monitored during transit. The AI system adjusts admixture additions or recommends minimal water additions to compensate for moisture variations. In the traditional truck, no adjustments were made. Upon arrival, the slump, air content, and compressive strength of the concrete were measured.
The reference is made to
A t-test was used in statistical analysis to compare the average slump, air content, and strength between the two systems for each moisture condition. The t-test results (p<0.01 for dry aggregate, p<0.05 for wet aggregate) show that the AI-controlled system produces concrete with significantly more consistent slump despite variations in aggregate moisture. This demonstrates a non-random advantage in handling inconsistencies in raw materials. In other words, the statistically significant improvement in slump consistency achieved by the present invention highlights its non-obvious ability to compensate for such variations and ensure consistent concrete quality.
Prior art, which mentions various sensors but does not specify their combined use or data fusion, does not teach a system capable of adapting to aggregate moisture variations in real-time. This experiment demonstrates the ability of the AI system of the present invention to compensate for variations in raw materials, ensuring consistent concrete quality, which is not suggested in the prior art. This leads to improved quality control and reduces the need for costly adjustments or material rejection at the construction site.
A series of 20 concrete deliveries using both the AI-controlled and traditional systems were conducted under various conditions (different mix designs, weather, distances). The type and quantity of each admixture used in both systems were meticulously tracked for each delivery. Also, tracked was the amount of water added at the construction site for each delivery.
AI-controlled system shows a statistically significant reduction (15-25%) in the total consumption of key admixtures (plasticisers, retarders, etc.) and water compared to the traditional system. Table 6 below summarises these obtained results.
A paired t-test was used in statistical analysis to compare the average admixture and water usage per cubic yard of concrete between the two systems across all deliveries. The statistically significant reduction (p<0.001) in plasticizer usage, and similar reductions (not shown here, but available upon request) in other admixtures and water, demonstrates a non-random advantage of the AI-controlled system in optimizing resource consumption.
This experiment demonstrates the surprising efficiency of the proactive AI system of the present invention in achieving desired concrete properties with less reliance on admixtures and water. This is a non-obvious outcome not suggested by the prior art, which focuses on initial optimisation or reactive adjustments. WO 2022/249162 A1, which is the publication of the co-pending application by the present inventors, mentions the use of historical data in AI to optimise initial batch proportions, but does not suggest a system capable of achieving such significant reductions in admixture and water usage through real-time, AI-driven control. The present result highlights the non-obvious efficiency of the invention in achieving desired concrete properties with fewer resources. This translates to significant cost savings due to reduced material consumption and promotes environmental sustainability by minimising waste and the use of potentially harmful chemicals.
Two sets of concrete test specimens (cylinders, beams) using concrete produced with the AI-controlled and traditional systems are constructed. The specimens are cured under standard conditions. Compressive strength tests are conducted at various ages (7 days, 28 days, and later ages). The specimens are monitored for cracking, durability, and other long-term performance indicators.
AI-controlled concrete was found to exhibit comparable or even superior long-term strength and durability despite potentially using less cement and admixtures compared to the traditional method. Table 7 below summarises these obtained results.
A t-test was used in statistical analysis to compare the average compressive strength at each age between the two sets of specimens. While the difference in 28-day strength might not be statistically significant in this simulated example, the comparable performance despite using less cement and admixtures is still noteworthy.
This experiment demonstrates the surprising ability of the AI system to optimise concrete mix proportions and admixture usage without compromising long-term performance. This is a non-obvious outcome not suggested by the prior art. None of the prior art references explicitly address the long-term performance implications of real-time admixture control. The ability of the present invention to reduce cement and admixture usage without compromising long-term performance is a non-obvious benefit that contributes to sustainability and cost-effectiveness. Indeed, it potentially reduced cement consumption, which is a major contributor to CO2 emissions while maintaining or even enhancing the durability and service life of concrete structures.
The statistically significant results from these experiments build a strong case for the non-obviousness of the present invention. The invention's ability to actively counteract temperature effects, prevent delays, adapt to raw material variations, optimise resource usage, and potentially enhance long-term performance while reducing reliance on cement and admixtures are all non-obvious benefits not suggested by the prior art.
Below are several examples of the concrete production process using the system of the present invention, during the transportation (conditions, actions, and possible reactions).
The experiment demonstrated the AI-controlled production of concrete with a target slump of 100 mm and a setting time of 4 hours. The ambient temperature is 25° C. The results of the experiment are summarised in the following table.
The reference is made to
As seen from the obtained results, the combination of retarders and water reducers helps control setting time and workability without excessive water addition. Continuous monitoring and data-driven decision-making enable proactive adjustments, ensuring optimal concrete quality. The proactive AI system is capable of considering even a wider range of factors and make more nuanced adjustments based on complex interactions between concrete components and environmental conditions.
In the present example, the aim was a target slump of 100 mm. The initial slump after mixing at the plant was found to be slightly below the target (around 90 mm). The concrete needed to maintain its workability for at least 2 hours during transportation. The ambient temperature is 25° C. The results of the experiment are summarised in the following table.
The reference is made to
The table below illustrates how the AI system in the concrete mixer truck monitors and manages concrete properties during transportation to prevent segregation and bleeding.
The following observations and actions were taken by the proactive AI system of the invention:
Thus, the AI-based system successfully maintained the concrete's properties within the desired range through timely adjustments of admixtures. The combination of retarders and water reducers helped control setting time and workability without excessive water addition. Continuous monitoring and data-driven decision-making enabled proactive adjustments, ensuring optimal concrete quality.
This example demonstrates the proactive nature of the AI system in preventing concrete quality issues during transportation. By continuously monitoring and making timely adjustments, it ensures that the concrete arrives at the construction site in optimal condition, minimising the risk of segregation and bleeding.
This experiment focuses on the proactive AI system's ability to regulate air entrainment in the concrete mixture in real time. Air entrainment is crucial for enhancing concrete's durability, particularly in freeze-thaw cycles. The system monitors air content throughout the process and makes adjustments as needed. The table below illustrates how the AI system in the concrete mixer truck monitors and manages concrete properties during transportation to prevent segregation and bleeding.
The following observations and actions were taken by the proactive AI system of the invention:
This experiment demonstrates the effectiveness of the proactive AI-powered system in controlling air entrainment in fresh concrete. The results of this experiment provide clear support for the proactive AI system's ability to continuously monitor air content, make real-time adjustments by adding more admixtures and maintain the desired air content despite variations during transportation. This ensures that the concrete delivered to the construction site has the required properties for durability and workability.
This experiment is designed to test the proactive AI-driven concrete production system's ability to adjust the setting time of concrete using an accelerator admixture. In this experiment, concrete needs to be delivered to a construction site with specific setting time and strength requirements, taking into account the influence of ambient temperature. The table below shows how the proactive AI-controlled system adjusts the concrete mixture during production and transportation to ensure it arrives at the construction site with the desired properties. The first column indicates the time elapsed since the beginning of the concrete production process. “Setting time” indicates the time it takes for the concrete to harden. “Admixture Action” describes any actions taken by the proactive AI system to adjust the concrete mixture, such as adding an accelerator or a water reducer. “Observations” provides explanations for the changes in concrete properties and the actions taken by the AI system.
In essence, the table outlines the decision-making process of the proactive AI system in maintaining the quality of the concrete throughout its production and transportation. It demonstrates how the system monitors various parameters, detects deviations from the desired range, and takes corrective actions by adding chemical admixtures as needed.
The results of this experiment show how the proactive AI system adapts to changing conditions, specifically the decreasing ambient temperature, to maintain the desired concrete properties. The initial setting time and strength of the concrete meet the requirements. However, as the ambient temperature drops, the setting time is delayed. To counteract this, the system adds an accelerator, which speeds up the setting process. The addition of the accelerator can sometimes affect the concrete's workability, so a water reducer is also added to ensure it remains easy to place.
This experiment clearly demonstrates the proactive AI system's ability to monitor relevant environmental and concrete properties, proactively adjust the concrete mix using admixtures, and maintain desired setting times and strength even with changing ambient temperatures. By using the proactive AI-driven system of the present invention, concrete producers can ensure consistent quality and meet specific construction requirements, even in challenging conditions.
This example demonstrates how the proactive AI of the invention fuses sensor data to manage concrete production. In this example, a concrete mixer truck of the invention was en route to a construction site on a hot day (30° C.). The concrete mix was prepared for a 4-hour setting time and a slump of 100 mm. Sensors used are: temperature sensor that monitored the concrete temperature inside the mixer, FLIR thermal imaging camera that captures thermal images of the concrete to assess its consistency and slump, and acoustic sensor that listens to the sounds of the mixing process to detect changes in viscosity.
The sensor data fusion and corresponding actions performed by the proactive AI of the invention are summarised in the following table.
As the concrete travels, the temperature increases, causing the slump to decrease and the viscosity to increase. At 60 minutes, the proactive AI detects these changes and adds a water reducer to increase workability without affecting setting time. At 120 minutes, the proactive AI notices further slump decrease and a rapid increase in viscosity, indicating potential early setting. It adds a retarder to slow down the hydration process and prevent premature stiffening. The proactive AI continues to monitor the concrete and takes no further action as the concrete properties stabilise within the desired range. The outcome of this is that the concrete arrives at the construction site with the desired workability and setting time, despite the challenging conditions.
This example demonstrates how the AI system fuses data from multiple sensors to provide a comprehensive understanding of the concrete's state. By analysing this data in real-time, the AI can proactively adjust the mixture, ensuring consistent quality throughout the production and transportation process.
This application is a Continuation-in-part of U.S. patent application Ser. No. 18/514,023, filed Nov. 20, 2023, which is a Continuation-in-part of PCT Patent Application No. PCT/IL2022/050173 having International filing date of Feb. 14, 2022, which claims the benefit of priority of U.S. Provisional Patent Application No. 63/192,693, filed May 25, 2021, the contents of which are all incorporated herein by reference in their entirety.
| Number | Date | Country | |
|---|---|---|---|
| 63192693 | May 2021 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 18514023 | Nov 2023 | US |
| Child | 18939807 | US | |
| Parent | PCT/IL2022/050173 | Feb 2022 | WO |
| Child | 18514023 | US |
