Liquid intelligence detection system and method

Abstract

The application provides a liquid intelligent detection system and method, which comprises: a shadow effect device closely attached to the outer side of a container containing a to-be-detected liquid, which generates an electric signal carrying the characteristics of the to-be-detected liquid when irradiated by light from the opposite side; an Arduino development board which receives and processes the electric signal to obtain a digital electric signal; and an upper computer which adopts a liquid classification function and / or a liquid multi-feature detection model to detect the digital electric signal to obtain a liquid detection result of the to-be-detected liquid. In the system, the shadow effect device can realize light detection to generate an electric signal carrying the characteristics of the to-be-detected liquid, and then realize liquid detection of the to-be-detected liquid based on the electric signal. The shadow effect device is obtained by depositing a metal layer on a substrate, has low cost, can realize self-power supply, has low power consumption, does not need to be immersed in the to-be-detected liquid during detection, ensures the accuracy and reliability of the liquid detection result, and has a simple detection process, thereby greatly improving the detection efficiency.

Description

Technical Field

[0001] This invention relates to the field of detection and analysis technology, and in particular to a liquid intelligent detection system and method. Background Technology

[0002] Liquid detection technology is involved in various fields such as chemical production, petroleum storage and transportation, biomedicine, food and beverage, and water quality testing, and is one of the key technologies affecting the Earth's ecological environment. Among them, liquid level measurement is the process of real-time and accurate monitoring and detection of the liquid level in a container, while liquid identification includes the detection of characteristics such as the color and concentration of the liquid in the container.

[0003] Liquid detection encompasses the detection of liquid characteristics such as level, color, and concentration, and is involved in numerous fields of social development and technological advancement. Liquid probes are widely used in liquid detection, and can be categorized into contact and non-contact types based on their detection methods. Contact liquid probes rely on the physical or chemical properties of the liquid and depend on the direct contact of a high-precision instrument probe. The electrical signal of a contact liquid probe depends on the physical or chemical properties of the liquid, such as dielectric constant, buoyancy, hydrostatic pressure, density, viscosity, and temperature. However, in most scenarios, these detection devices are complex to install and deploy, sometimes requiring testing in specialized laboratories, and have high requirements for the detection environment. Secondly, contact liquid probes, such as capacitive, buoyant, and hydrostatic probes, require immersion in the liquid for measurement, leading to corrosion, contamination, or scaling of the probe and high maintenance costs. Furthermore, in dynamic liquid monitoring scenarios, contact liquid probes typically exhibit nonlinearity and hysteresis errors, requiring complex signal processing and compensation to achieve precise measurements.

[0004] In contrast to the problems associated with contact liquid probes, non-contact liquid probes have emerged, capable of detecting liquids based on wireless signals, optical signals, or photoelectric detection. Wireless-based non-contact liquid probes can calculate liquid properties such as level, dielectric constant, and density by transmitting and receiving signals from ultrasonic waves, radar, RFID (Radio Frequency Identification), and millisecond waves. The LiquID liquid probe designed by Dhekne et al. utilizes two independent ultra-wideband (UWB) units to measure the dielectric constant of liquids for liquid identification. Wang et al. developed the TagScan system using RFID signals, matching the properties of the liquid to be detected with a database for measurement. Other studies have used FG-LiquID and MmLiquid liquid probes to measure minute information regions in the millimeter-wave spectrum, extracting the most discriminative features of the liquid. However, these methods require prior knowledge of the specific properties of the liquid and necessitate complex and specialized signal transmitting devices to transmit UWB or RFID signals.

[0005] Optical signal-based liquid probe technology is another method for non-contact liquid detection, generally identifying and detecting liquids through further analysis of images captured by a camera. Yue et al. developed the CapCam liquid probe to detect liquids by measuring the surface tension of liquids using a mobile phone camera. Wu et al. used subtle viscosity differences in liquid images captured by a camera to identify liquids. However, optical signal-based liquid detection has limitations related to the environment and shooting angle. For example, in low light conditions or when certain angles are obstructed, liquid features cannot be captured by the camera, limiting its application and development. On the other hand, this probe technology significantly increases computational power consumption and energy consumption due to the high precision requirements of the camera and the complexity of the image data.

[0006] Non-contact liquid detection based on photoelectric detection is another means to achieve intelligent liquid detection, with advantages such as short response time, high accuracy, and stability. Recent studies have utilized the unique absorption spectrum of liquids to specific light sources (such as infrared light) to acquire optical absorption or reflection information of liquids using photoelectric probes. The Nutrilyzer liquid probe developed by Rahmman et al. uses absorption spectroscopy to achieve liquid identification, capturing the transmission and scattering characteristics of LED light in liquid food to detect nutrients or impurities. However, this probe requires sampling the liquid onto the electrodes of a piezoelectric detector, making the detection process relatively cumbersome. Another study proposed an intelligent device equipped with a near-infrared light source to estimate alcohol concentration levels. While photoelectric detection using a specific light source can protect liquid detection from ambient light, the additional light source increases equipment cost and energy consumption. Furthermore, these photoelectric detection liquid probes cannot completely solve the problems of ambient light interference and low resolution in low light conditions, and their limitation to specific liquid identification also restricts their further development. Therefore, developing non-contact liquid probes with excellent detection capabilities in low ambient light is crucial.

[0007] In summary, existing liquid detection systems cannot simultaneously achieve the characteristics of high precision, low cost, low energy consumption, and high efficiency. Summary of the Invention

[0008] In view of this, the purpose of the present invention is to provide a liquid intelligent detection system and method to alleviate the technical problems of existing liquid detection systems being unable to simultaneously achieve high precision, low cost, low energy consumption and high efficiency.

[0009] In a first aspect, embodiments of the present invention provide a liquid intelligent detection system, comprising: a shadow effect device, an Arduino development board, and a host computer connected in sequence, wherein the shadow effect device is obtained by depositing a metal layer on a substrate;

[0010] The shadow effect device is attached to the outside of the container containing the liquid to be tested, and is used to generate an electrical signal carrying the characteristics of the liquid to be tested when illuminated by light from the opposite side. The electrical signal is formed by the difference in work function between the illumination area and the shadow area of ​​the shadow effect device. The illumination area is the area formed after the light has not passed through the liquid to be tested, and the shadow area is the area formed after the light has passed through the liquid to be tested.

[0011] The Arduino development board is used to receive the electrical signal and process the electrical signal to obtain a digital electrical signal;

[0012] The host computer is used to perform liquid detection on the digital electrical signal using a liquid classification function and / or a liquid multi-feature detection model to obtain the liquid detection result of the liquid to be tested.

[0013] Furthermore, the shadow effect device includes a silicon substrate and a metal layer deposited on the silicon substrate, wherein the metal layer is not deposited at the middle position of the silicon substrate.

[0014] Furthermore, the metal of the metal layer includes: gold;

[0015] The width of the undeposited metal layer is a preset width.

[0016] Furthermore, the processing includes filtering and analog-to-digital conversion.

[0017] Furthermore, the liquid classification function includes any one of the following: color classification function, liquid level classification function, and concentration classification function.

[0018] Furthermore, the liquid multi-feature detection model is a 1D convolutional neural network model;

[0019] The liquid detection results include: liquid color information, liquid level information, and liquid concentration information.

[0020] Furthermore, the host computer is also used to visualize the liquid detection results.

[0021] Furthermore, the number of shadow effect devices is 1 or 2. When the number of shadow effect devices is 2, the shadow effect devices are placed side by side and closely attached to the outside of the container containing the liquid to be tested.

[0022] Furthermore, the liquid multi-feature detection model is pre-trained.

[0023] Secondly, embodiments of the present invention also provide a liquid intelligent detection method, applied to the liquid intelligent detection system described in any of the first aspects above, the method comprising:

[0024] When a shadow effect device is placed close to the outside of a container containing a liquid to be tested, it generates an electrical signal carrying the characteristics of the liquid to be tested when illuminated by light from the opposite side. The electrical signal is formed by the difference in work function between the illumination area and the shadow area of ​​the shadow effect device. The illumination area is the area formed after the light has not passed through the liquid to be tested, and the shadow area is the area formed after the light has passed through the liquid to be tested.

[0025] The Arduino development board receives the electrical signal and processes it to obtain a digital electrical signal.

[0026] The host computer uses a liquid classification function and / or a liquid multi-feature detection model to perform liquid detection on the digital electrical signal, and obtains the liquid detection result of the liquid to be tested.

[0027] In this embodiment of the invention, a liquid intelligent detection system is provided, comprising: a shadow effect device, an Arduino development board, and a host computer connected in sequence. The shadow effect device is obtained by depositing a metal layer on a substrate. The shadow effect device is attached to the outside of a container holding the liquid to be tested, and is used to generate an electrical signal carrying the characteristics of the liquid to be tested when illuminated by light from the opposite side. The electrical signal is formed by the difference in work function between the illuminated area and the shadowed area of ​​the shadow effect device. The illuminated area is the area formed before the light passes through the liquid to be tested, and the shadowed area is the area formed after the light passes through the liquid to be tested. The Arduino development board is used to receive the electrical signal and process it to obtain a digital electrical signal. The host computer is used to perform liquid detection on the digital electrical signal using a liquid classification function and / or a liquid multi-feature detection model to obtain the liquid detection result of the liquid to be tested. As described above, in the intelligent liquid detection system of the present invention, the shadow effect device can realize light detection, thereby generating an electrical signal carrying the characteristics of the liquid to be tested, and then realizing liquid detection of the liquid to be tested based on the electrical signal. The shadow effect device is obtained by depositing a metal layer on a substrate, which is low in cost and can be self-powered with low power consumption. During detection, the shadow effect device does not need to be immersed in the liquid to be tested, which ensures the accuracy and reliability of the liquid detection results. The detection process is simple, which greatly improves the detection efficiency and alleviates the technical problem that existing liquid detection systems cannot simultaneously achieve high precision, low cost, low energy consumption and high efficiency. Attached Figure Description

[0028] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0029] Figure 1 This is a schematic diagram of the structure of a liquid intelligent detection system provided in an embodiment of the present invention;

[0030] Figure 2 A schematic diagram of the shadow effect device provided in an embodiment of the present invention;

[0031] Figure 3 A schematic diagram illustrating the working mechanism of the shadow effect device provided in an embodiment of the present invention;

[0032] Figure 4 A schematic diagram of the structure of the dual-channel intelligent liquid detection system provided in an embodiment of the present invention;

[0033] Figure 5 A schematic diagram illustrating the visualized liquid detection results provided in an embodiment of the present invention;

[0034] Figure 6 A flowchart of a liquid intelligent detection method provided in an embodiment of the present invention;

[0035] Figure 7 A schematic diagram illustrating the output characteristics and stability of the shadow effect device provided in an embodiment of the present invention. Detailed Implementation

[0036] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Existing liquid detection systems cannot simultaneously achieve the characteristics of high precision, low cost, low energy consumption, and high efficiency.

[0038] Based on this, in the intelligent liquid detection system of the present invention, the shadow effect device can realize light detection, thereby generating an electrical signal carrying the characteristics of the liquid to be tested, and then realizing liquid detection of the liquid to be tested based on the electrical signal. The shadow effect device is obtained by depositing a metal layer on a substrate, which is low in cost and can be self-powered with low power consumption. During detection, the shadow effect device does not need to be immersed in the liquid to be tested, which ensures the accuracy and reliability of the liquid detection results. The detection process is simple and greatly improves the detection efficiency.

[0039] To facilitate understanding of this embodiment, a detailed description of a liquid intelligent detection system disclosed in this embodiment of the invention will be provided first.

[0040] Example 1:

[0041] Figure 1 This is a schematic diagram of the structure of a liquid intelligent detection system according to an embodiment of the present invention, as shown below. Figure 1 As shown, it includes: a shadow effect device, an Arduino development board, and a host computer connected in sequence, wherein the shadow effect device is obtained by depositing a metal layer on a substrate;

[0042] The shadow effect device is placed close to the outside of the container containing the liquid to be tested. It is used to generate an electrical signal carrying the characteristics of the liquid to be tested when illuminated by light from the opposite side. The electrical signal is formed by the difference in work function between the illumination area and the shadow area of ​​the shadow effect device. The illumination area is the area formed after the light has not passed through the liquid to be tested, and the shadow area is the area formed after the light has passed through the liquid to be tested.

[0043] An Arduino development board is used to receive electrical signals and process them to obtain digital electrical signals.

[0044] The host computer is used to perform liquid detection on digital electrical signals using liquid classification functions and / or liquid multi-feature detection models to obtain the liquid detection results of the liquid to be tested.

[0045] Specifically, shadows are ubiquitous in production and daily life. When light passes through a liquid being tested, it projects the liquid's shadow. Past research neglected the fact that analyzing the shadow information of the liquid can reveal various characteristics such as color, concentration, and liquid level. The inventors previously discovered that a metal thin-film-semiconductor Schottky junction can optically modulate the work function of the surface metal by projecting local shadows (defined as the "shadow effect"). Based on this new optical physics phenomenon, the inventors developed a novel optoelectronic device beyond traditional optoelectronic devices. This device utilizes the shadow falling on it to create a light intensity difference, causing a work function difference between the light and dark parts of the Schottky junction. This creates a potential barrier that forces electrons to flow from the bright side to the dark side, generating a current in the circuit when connected to an external circuit. This novel shadow effect optoelectronic device (referred to as the shadow effect device) has no complex design structure, is simple to prepare materials for, and exhibits stable performance.

[0046] The rapid development of computers has driven the intelligentization of monitoring and detection technologies. Detection devices achieve intelligent automatic control and management through the combination of control systems and main control chips. Liquid detection devices need to accurately identify multiple characteristics of liquids, especially in dynamic environments, where it is extremely difficult to directly and manually quantitatively identify the comprehensive characteristics of liquids using output signals. To enhance the accuracy of liquid detection devices in dynamic environments, it is necessary to integrate the detected signal data with deep learning (DL) technology. As an emerging subfield of machine learning, DL has been widely used in smart wearables, healthcare, and many other fields. Research has shown that liquid detection devices can extract subtle key features from signals through DL, effectively classify electrical signals, and achieve a higher level of accuracy. Especially for non-contact liquid detection signals based on photoelectric detection, the basic features of the signal can be extracted under the analysis of DL algorithms, reducing interference signals. Therefore, combining shadow effect liquid probes with DL algorithms for signal recognition can accurately identify liquid characteristics while achieving low cost and convenience, promoting the practical application of shadow effect-based intelligent photoelectric detection systems in the field of liquid identification.

[0047] Specifically, a shadow effect device is used as a photodetector; by connecting to an Arduino development board, the analog voltage signal is converted into a one-dimensional digital signal via an ADC; the system interface and functions are developed based on Python, and different digital electrical signals are identified through liquid classification functions and / or liquid multi-feature detection models, enabling multifunctional real-time intelligent liquid detection, including liquid color, liquid concentration, and liquid level detection. The aforementioned light can be ambient light.

[0048] The aforementioned shadow effect device serves as a novel method for liquid detection. The shadow cast by the liquid can be interpreted as a collective result of light absorption, scattering, transmission, and reflection after passing through the liquid. Therefore, shadows offer a potential non-contact method to provide information about the detection of the liquid.

[0049] The application design concept of the intelligent liquid detection system of this invention is based on the shadow effect performance. By measuring the ambient light occlusion of the shadow effect device along one dimension, it quantifies and identifies translucent or opaque test liquids in open or transparent containers. Specifically, when the test liquid is poured into the container, as the volume of liquid in the container increases, pixels submerged below or near the liquid surface will receive less light, and the shadow area of ​​the shadow effect device will increase accordingly. The increase in the shadow area of ​​the shadow effect device leads to a change in the photoelectric signal, which can be used to infer the liquid level in the container. Secondly, when the shadow area is fixed, the change in the contrast between the shadow part and the illuminated part of the shadow effect device will also cause different output performance of the shadow effect device, which can be used to detect the liquid concentration in the container. In addition, since the shadow effect device responds differently to different wavelengths of light (different colored liquids absorb light differently, resulting in different wavelength responses), in practical applications, it can be used to identify the color of different liquids.

[0050] In this embodiment of the invention, a liquid intelligent detection system is provided, comprising: a shadow effect device, an Arduino development board, and a host computer connected in sequence. The shadow effect device is obtained by depositing a metal layer on a substrate. The shadow effect device is attached to the outside of a container holding the liquid to be tested, and is used to generate an electrical signal carrying the characteristics of the liquid to be tested when illuminated by light from the opposite side. The electrical signal is formed by the difference in work function between the illuminated area and the shadowed area of ​​the shadow effect device. The illuminated area is the area formed before the light passes through the liquid to be tested, and the shadowed area is the area formed after the light passes through the liquid to be tested. The Arduino development board is used to receive the electrical signal and process it to obtain a digital electrical signal. The host computer is used to perform liquid detection on the digital electrical signal using a liquid classification function and / or a liquid multi-feature detection model to obtain the liquid detection result of the liquid to be tested. As described above, in the intelligent liquid detection system of the present invention, the shadow effect device can realize light detection, thereby generating an electrical signal carrying the characteristics of the liquid to be tested, and then realizing liquid detection of the liquid to be tested based on the electrical signal. The shadow effect device is obtained by depositing a metal layer on a substrate, which is low in cost and can be self-powered with low power consumption. During detection, the shadow effect device does not need to be immersed in the liquid to be tested, which ensures the accuracy and reliability of the liquid detection results. The detection process is simple, which greatly improves the detection efficiency and alleviates the technical problem that existing liquid detection systems cannot simultaneously achieve high precision, low cost, low energy consumption and high efficiency.

[0051] The above provides a brief overview of the intelligent liquid detection system of the present invention. The specific details involved are described in detail below.

[0052] In an optional embodiment of the present invention, the shadow effect device includes a silicon substrate and a metal layer deposited on the silicon substrate, wherein no metal layer is deposited at the middle position of the silicon substrate.

[0053] In an optional embodiment of the present invention, the metal of the metal layer includes: gold;

[0054] The width of the undeposited metal layer is a preset width.

[0055] The fabrication and working principle of shadow effect devices are introduced below:

[0056] Shadow effect devices are fabricated by depositing Au thin films on n-type silicon substrates. These devices utilize the difference in work function between the illuminated and shaded regions to generate in-plane charge transport, resulting in significant electrical output and potential difference. This process offers advantages such as low cost and ease of fabrication. Shadow effect devices exhibit high resolution and high stability even under weak ambient light. Furthermore, they are self-powered, allowing low-power systems to achieve self-sustaining operation through energy harvesting.

[0057] Fabrication process of shadow effect device: A positive silicon wafer is cut into small pieces of a specified size, and then rinsed successively with ethanol, isopropanol, and water. Afterwards, the silicon substrate is dried with nitrogen gas. A copper tape of a predetermined width is attached to the center of the silicon wafer as a channel. Au is then deposited at a thickness of 7-100 nm using a thermal evaporation method. Finally, the copper tape is removed, resulting in the shadow effect device. Figure 2 As shown.

[0058] The working mechanism of shadow effect devices is as follows Figure 3 When the shadow effect device is placed in the shadow region, due to the work function of Au (W) m When the energy level is greater than n-Si, free electrons will flow from Si to Au until the Fermi level is equal. In equilibrium, a Schottky barrier (W) will form at the Au / Si interface. b When the shadow effect device is fully placed in the illuminated area, the absorption of photons with energy above the band gap generates electron-hole pairs through transitions. Under the influence of the internal electric field, free carriers move in different directions: electrons move towards n-Si, and holes move towards Au. The accumulated electrons overcome the Schottky barrier (W). b * And inject metal. When half of the shadow effect device is blocked and the other half is exposed to light, the work function (W) of Au in the illuminated area is... m * The work function (W) of Au above the shaded region m When a shadow effect device is connected to an external circuit, it drives free electrons to flow from the illuminated portion of the Au film to the shaded portion, forming a continuous current. To make electron transfer more efficient, the middle portion of the Au film (i.e., the middle position of the silicon substrate where no metal layer is deposited) is removed during the fabrication of the shadow effect device. This allows electrons on the Au film surface to flow only from the external circuit, preventing electrons from flowing directly from the Au film to the shaded area.

[0059] In an optional embodiment of the present invention, the processing includes filtering and analog-to-digital conversion.

[0060] In an optional embodiment of the present invention, the host computer is also used to visualize the liquid detection results.

[0061] The host computer system interface is developed using Python to create a display interface for the liquid detection system. The signal conversion program is burned onto an Arduino development board. The host computer receives digital electrical signals, which are then transmitted to Python. A single-channel signal (e.g., ...) is used. Figure 1 As shown, with a shadow effect device quantity of 1, it is possible to detect the color and concentration of liquids using dual-channel signals (such as...). Figure 4As shown, the number of shadow effect devices is 2, and the two shadow effect devices are set side by side at different heights. This can realize the detection of liquid level and multiple features (three features at the same time). (Because the shadow effect device has the maximum response when the shadow area occupies 50% of the entire device, and the response decreases when it is greater than 50% or less than 50%. If the voltage at 50% is 2V, the actual voltage detected by one shadow effect device is 1V. It is impossible to determine whether the shadow area corresponding to the 1V voltage is above or below 50%, so it is impossible to determine the liquid level. Therefore, two shadow effect devices are needed to cooperate in detection. They are set side by side at different heights. Specifically, the bottom of one can be located in the middle of the other.)

[0062] The Python-based classification models include: simple single-feature classification (i.e., liquid classification functions (probability distribution functions), which can be any of the following: color classification function, liquid level classification function, and concentration classification function; the resulting liquid detection results have the same functionality as the corresponding liquid classification function) and multi-feature classification based on machine learning algorithms (i.e., liquid multi-feature detection models), which classify digital voltage signals using these models. Finally, the Matplotlib plotting library is used to create a visualization interface to display the liquid detection results, including liquid color, concentration, and liquid level information, such as... Figure 5 As shown.

[0063] In an optional embodiment of the present invention, the liquid multi-feature detection model is: a 1D convolutional neural network model;

[0064] Liquid detection results include: liquid color information, liquid level information, and liquid concentration information;

[0065] The liquid multi-feature detection model was obtained through pre-training.

[0066] Specifically, the shadow effect device acts as a photodetector, connecting to an Arduino development board to convert analog electrical signals into one-dimensional digital signals via an ADC. Voltage values ​​are acquired by a host computer, the data is normalized, and converted into a matrix-style dataset. Labels are then added to the dataset to categorize it, dividing the dataset into a training set and a prediction set.

[0067] Deep learning relies on a central processing unit (CPU) or graphics processing unit (GPU) for training, using the Keras architecture based on TensorFlow, and employing a 1D convolutional neural network (CNN) model. The algorithm automatically divides the dataset into training and test sets, using deep feature extraction from the training set for training and electrical signal classification; the test set is used to check the model's accuracy, and appropriate model parameters are saved for detecting new signals, such as... Figure 4 As shown.

[0068] The present invention has the following advantages:

[0069] 1. Shadow effect devices, as photodetectors, bring a new perspective to liquid detection. Shadows are ubiquitous, but are often overlooked in optoelectronic applications. This invention, based on the principle of the shadow effect, studies the shadow cast by a liquid as a novel method for liquid detection and identification. Different liquids produce different shadows. By constructing a non-contact liquid probe using the shadow effect principle, liquid features can be detected through shadow differences. That is, by utilizing the shadow contrast caused by liquid features, ambient light detection can be achieved without the need for an additional light source. Even in weak ambient light, it has high resolution and high stability. This can reduce the energy consumption and complexity of the detection system, and improve its adaptability and flexibility.

[0070] 2. Shadow effect devices do not require immersion in liquid (i.e., non-contact liquid detection) or an additional light source, improving their resistance to harsh environmental conditions. They are suitable for detecting various types and states of liquids, such as transparent or opaque, monochromatic or multicolor, stable or dynamic, and significantly reduce subsequent maintenance costs. Compared with existing technologies, while ensuring that the detection results are not affected by the properties of the liquid itself, it also avoids contamination or damage to the liquid, thus guaranteeing the accuracy and reliability of the detection results, and extending the service life and stability of the detection system.

[0071] 3. By analyzing and identifying the acquired one-dimensional digital signals, multifunctional liquid detection can be achieved, such as liquid color, liquid concentration, and liquid level detection. This invention can select appropriate machine learning models and parameters according to different detection needs and scenarios (e.g., the machine learning models and parameters selected will differ when detecting water quality and when detecting chemicals), thereby improving detection efficiency and accuracy. Compared with existing technologies, while ensuring the diversity and richness of detection results, it also simplifies the detection process and operating steps.

[0072] 4. Shadow effect devices are simple to manufacture, offering advantages such as low cost and ease of production. This reduces the initial investment and operating costs of the detection system, improving its economic viability and widespread adoption.

[0073] 5. Shadow effect devices can be self-powered, and low-power systems can achieve self-sustaining operation through energy harvesting, making them applicable to a wider range of scenarios. This increases the mobility and portability of the detection system, improves its adaptability and flexibility, and expands its application scope and fields.

[0074] 6. It possesses intelligent features, enabling automatic adjustment and feedback based on real-time detection results, such as adjusting the sensitivity and position of shadow effect devices, displaying or storing detection data, and triggering alarms or controlling related equipment. Furthermore, this invention can also be linked with other intelligent devices or networks to achieve remote monitoring and management. Compared to existing technologies, while ensuring the real-time nature and reliability of detection results, it also improves the intelligence level and application scope of the detection system.

[0075] 7. It has the advantages of low cost, low energy consumption, easy installation and easy maintenance, and can be widely used in liquid detection in various fields to improve production efficiency and quality and protect people's health and safety;

[0076] 8. Liquid detection is achieved by utilizing the varying areas created by the device's shadow, illumination contrast, and response to different wavelengths of light. The sensor signal is converted into a digital signal via hardware circuitry. The detection system algorithm, combined with a deep learning neural network architecture, can extract effective signal features. Combined with the inherent advantages of the device, this results in higher detection accuracy. It effectively addresses the high resolution and high power consumption requirements of image sensing and is expected to simultaneously detect multiple liquid characteristic parameters. This enhances the functionality and intelligence of the detection system, meeting the diverse needs of different users and scenarios.

[0077] The intelligent liquid detection system of this invention has a wide range of practical applications and promising prospects. It can be applied to fields such as industrial production, automated medical monitoring, transportation, and environmental monitoring, providing an efficient, accurate, and convenient solution for the detection and control of various liquids. This intelligent liquid detection system utilizes photoelectric sensors to perform multi-dimensional detection of liquids, including parameters such as liquid color, concentration, and level. Through the identification and processing of different signals, real-time intelligent identification and monitoring of liquids can be achieved.

[0078] In the field of industrial production, the intelligent liquid detection system of this invention can be used for liquid level measurement in reaction vessels or storage tanks of chemical, petroleum, and mining enterprises to ensure automated and safe production and automated equipment management; it can also be used for water level measurement in reservoirs, ports, or waterways of hydropower, port, and waterway departments to improve water resource utilization efficiency and flood and drought prevention capabilities; it can also be used for oil level measurement in the fuel tanks of aircraft, tanks, and other power equipment of defense departments to ensure normal operation and combat readiness of equipment.

[0079] In the field of automated medical monitoring, the intelligent liquid detection system of this invention can be used for monitoring and controlling the level of intravenous infusion in patients to avoid adverse effects caused by infusion being too fast or too slow; it can also be used for monitoring and controlling the blood volume of venous blood draws to reduce the workload and errors of medical staff; it can also be used for monitoring and controlling the liquid storage or status in medical equipment such as artificial kidney machines, automatic gastric lavage machines, and medicated bath machines, and can automatically and manually replenish fluids in a timely manner when fluids are insufficient to ensure the safe operation of medical equipment and the therapeutic effect.

[0080] In the field of transportation, the intelligent liquid detection system of the present invention can be used for the rapid identification and authentication of liquid items in baggage security checks at airports, train stations, subways, etc., to improve the efficiency and accuracy of security checks; it can also be used for the monitoring and control of important liquid parameters such as fuel tanks or coolant in vehicles such as cars and trains, to improve the performance and safety of vehicles.

[0081] In food safety, this invention can be used to quickly identify the color and concentration of liquid products such as wine, juice, detergents, and vinegar to ensure product quality and consumer health. In environmental protection, this invention can be used for preliminary online monitoring and evaluation of the color and concentration of surface water to promote the protection and management of water resources. The liquid level identification function can also be applied to the monitoring of water levels in reservoirs, dams, or rivers.

[0082] In summary, this invention is a liquid intelligent detection system that combines photoelectric detectors. It features multifunctionality, high sensitivity, high stability, low cost, real-time operation, and intelligence. It is suitable for various occasions that require liquid detection and control, and has great practical value and application potential.

[0083] This invention brings a brand-new research approach to liquid intelligent detection technology, providing a non-contact liquid detection technology for important fields such as water quality detection, biomedicine, high-throughput synthesis, and military aerospace. It also provides a research foundation and platform for the continuous development of new optoelectronic devices and intelligent detection systems, promoting the cross-integration research of liquid detection and new optoelectronic devices.

[0084] This invention enables non-contact liquid detection in low-light environments using shadow effect devices. By studying the functional relationship between the shadow effect liquid probe and device size, the area of ​​the shadow cast by the liquid, and the response to different wavelengths, highly sensitive detection of non-specific liquid characteristics such as level, color, and concentration is achieved. Through back-end circuit chip design and the use of deep learning algorithms, a shadow effect liquid intelligent detection system is further constructed. The system design enables intelligent liquid shadow analysis, achieving low cost, zero power consumption, high resolution, and high sensitivity for real-time detection of multi-feature liquids in low-light environments. It also boasts advantages such as non-contact operation, low cost, self-driving, high sensitivity, and intelligence.

[0085] Example 2:

[0086] This invention also provides a liquid intelligent detection method, applied to the liquid intelligent detection system in Embodiment 1 above, with reference to... Figure 6 The method includes:

[0087] In step S602, after the shadow effect device is placed close to the outside of the container containing the liquid to be tested, when it is illuminated by light from the opposite side, it generates an electrical signal carrying the characteristics of the liquid to be tested. The electrical signal is formed by the difference in work function between the illumination area and the shadow area of ​​the shadow effect device. The illumination area is the area formed after the light has not passed through the liquid to be tested, and the shadow area is the area formed after the light has passed through the liquid to be tested.

[0088] In step S604, the Arduino development board receives the electrical signal and processes it to obtain a digital electrical signal.

[0089] Step S606: The host computer uses a liquid classification function and / or a liquid multi-feature detection model to perform liquid detection on the digital electrical signal to obtain the liquid detection result of the liquid to be tested.

[0090] The liquid detection mechanism analysis method of the present invention is as follows:

[0091] A non-specific liquid (such as water or ethanol) with different characteristic parameters (color, level, concentration) was used to project shadows in front of the shadow effect device, and the corresponding output voltage and current were measured using a Keithley 2400. By changing the liquid's color and concentration, the wavelength of light transmitted through the shadow and the shadow contrast were altered; by changing the liquid level, the shadow area and position were changed. The effects of different shadow wavelengths, shadow contrasts, shadow areas, and shadow positions projected by the non-specific liquid on the output of the shadow effect device were characterized and calculated. By comparing the relationship between illumination wavelength and shadow effect liquid probe parameters, the influence mechanism of the liquid's characteristic parameters on the output voltage and current of the shadow effect device was analyzed, and the corresponding functional relationship was established. The stability of the shadow effect device was repeatedly tested under ambient light with liquid circulation to meet the long-term stable service requirements of practical applications.

[0092] Figure 7 The output characteristics and stability of the shadow effect device are shown. (a) The relationship between the electrical performance of the shadow effect device and light intensity. (b) A comparison of Voc and Isc of the shadow effect device at different light wavelengths (450nm, 550nm, 650nm, 765nm, 850nm, and 940nm). (c) The effect of the contrast between illuminance of the shaded and illuminated parts on the performance of the shadow effect device. (d) The effect of the shadow area ratio (0%, 25%, 50%, 75%, and 100%) on the performance of the shadow effect device.

[0093] Explanation of the performance of shadow effect devices:

[0094] like Figure 7 As shown in Figure a, when the shadow area of ​​the shadow effect device is covered by black paper, Isc and Voc depend on light intensity and increase with increasing light intensity. This is especially true under ambient light conditions (below 100W m). -2 The subtle changes in light intensity lead to significant differences in device output. This is because the device's output signal depends on the illuminance contrast rather than the absolute illuminance, demonstrating the high resolution of the shadow effect device under low light illumination.

[0095] like Figure 7 As shown in b, the short-circuit current (Isc) and open-circuit voltage (Voc) of the shadow effect device with half-black paper shadows exhibit significant differences under different wavelengths of light. Therefore, under the same light source, the differences in Isc and Voc of the shadow effect device in the shadow and illuminated areas can be observed using Δλ (light wavelength variation). Moreover, since Δλ is a deviation, the shadow effect device does not require a specific emitter.

[0096] The light transmittance of a liquid is a key factor determining the brightness of its shadow. Using different layers of A4 white and black printing paper simulates liquids with varying transmittances, creating semi-shades of different brightness above them. For example... Figure 7 As shown in Figure c, the Isc and Voc of the shadow effect device increase continuously with the increase of the contrast between the illuminated and shadowed areas. Therefore, the Isc and Voc of the shadow effect device are negatively correlated with the liquid transmittance.

[0097] The shadow area is related to the height of the liquid. Similarly, by covering the shadow effect device with black paper to simulate the shadow produced by the liquid, the proportion of the shadow area to the total area can be changed by altering the position of the black paper above the shadow effect device. Figure 7 The results show that when the entire shadow effect device is fully illuminated (0%) or fully shaded (100%), both Isc and Voc decrease significantly. However, under half-shade (50%) conditions, Isc and Voc are higher than in other shadow areas (0%, 25%, 75%, and 100%). This phenomenon occurs due to the variation in shadow area.

[0098] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the method described above can be referred to the corresponding process in the aforementioned system embodiments, and will not be repeated here.

[0099] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0100] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0101] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A liquid intelligent detection system, characterized in that, include: The shadow effect device, Arduino development board, and host computer are connected in sequence. The shadow effect device includes: a silicon substrate and a metal layer deposited on the silicon substrate, wherein the metal layer is not deposited in the middle of the silicon substrate; the number of shadow effect devices is 2, and the two shadow effect devices are placed side by side and closely attached to the outside of the container containing the liquid to be tested, and the bottom of one shadow effect device is located in the middle of the other shadow effect device. The shadow effect device is attached to the outside of the container containing the liquid to be tested, and is used to generate an electrical signal carrying the characteristics of the liquid to be tested when illuminated by light from the opposite side. The electrical signal is formed by the difference in work function between the illumination area and the shadow area of ​​the shadow effect device. The illumination area is the area formed after the light has not passed through the liquid to be tested, and the shadow area is the area formed after the light has passed through the liquid to be tested. The Arduino development board is used to receive the electrical signal and process the electrical signal to obtain a digital electrical signal; The host computer is used to perform liquid detection on the digital electrical signal using a liquid classification function and / or a liquid multi-feature detection model to obtain the liquid detection result of the liquid to be tested, wherein the liquid multi-feature detection model is a 1-dimensional convolutional neural network model.

2. The intelligent liquid detection system according to claim 1, characterized in that, The metal in the metal layer includes: gold; The width of the undeposited metal layer is a preset width.

3. The intelligent liquid detection system according to claim 1, characterized in that, The processing includes filtering and analog-to-digital conversion.

4. The intelligent liquid detection system according to claim 1, characterized in that, The liquid classification function includes any one of the following: color classification function, liquid level classification function, and concentration classification function.

5. The intelligent liquid detection system according to claim 1, characterized in that, The liquid detection results include: liquid color information, liquid level information, and liquid concentration information.

6. The intelligent liquid detection system according to claim 1, characterized in that, The host computer is also used to visualize the liquid detection results.

7. The intelligent liquid detection system according to claim 5, characterized in that, The liquid multi-feature detection model was pre-trained.

8. A liquid intelligent detection method, characterized in that, The method applied to the liquid intelligent detection system according to any one of claims 1 to 7 includes: Two shadow effect devices are arranged side by side and closely attached to the outside of a container containing a liquid to be tested, with the bottom of one shadow effect device located in the middle of the other shadow effect device. Each shadow effect device includes a silicon substrate and a metal layer deposited on the silicon substrate, wherein the metal layer is not deposited in the middle of the silicon substrate. When the shadow effect device is illuminated by light from the opposite side, it generates an electrical signal carrying the characteristics of the liquid under test. The electrical signal is formed by the difference in work function between the illumination area and the shadow area of ​​the shadow effect device. The illumination area is the area formed after the light has not passed through the liquid under test, and the shadow area is the area formed after the light has passed through the liquid under test. The Arduino development board receives the electrical signal and processes it to obtain a digital electrical signal. The host computer uses a liquid classification function and / or a 1D convolutional neural network model as a liquid multi-feature detection model to perform liquid detection on the digital electrical signal, and obtains the liquid detection result of the liquid to be tested.

Images