Smart water dispensing device for direct drinking water terminals

Abstract

This intelligent water dispensing device for direct drinking water terminals solves the problems of multi-type water dispensing and intelligent sterilization of pipe networks in piped direct drinking water terminals, belonging to the field of terminal water dispensing technology. The invention includes a sampling pool, a rapid heating unit, a cooling unit, and a water dispensing unit on the supply and return water pipelines. The control unit controls the operation of the rapid heating unit, cooling unit, and water dispensing unit according to instructions input from the human-machine interface unit. A Fourier layered illumination source is incident on the sampling pool at staggered times, and an optical sensor receives the diffracted light generated in the sampling pool at different times, obtaining low spatial resolution three-dimensional images at each moment. The control unit controls the sampling pool to sample water in the supply and return water pipelines, then controls the Fourier layered illumination source and optical sensor to reconstruct a high spatial resolution image, determines whether sterilization is needed, and feeds back the sterilization signal to the upper control module to control the front-end central supply and return water device to perform overall sterilization of the supply and return water system, ensuring water safety from the source.

Description

Technical Field

[0001] This invention relates to an intelligent water dispensing device for direct drinking water terminals, belonging to the field of terminal water dispensing technology. Background Technology

[0002] With the continuous development of society and the economy, people's pursuit of quality of life is getting higher and higher. The drinking water industry urgently needs transformation and upgrading. The construction of urban piped drinking water systems can provide urban residents with more reliable and convenient drinking water, increase people's well-being, and effectively safeguard drinking water safety. However, due to the instability of piped drinking water quality, the problem of bacteria and microorganisms in drinking water equipment has always been a focus of attention.

[0003] Monitoring of domestic water mainly includes detecting chemical elements in samples, such as COD, TOC, ammonia nitrogen, fluoride, chloride, nitrate, nitrite, sulfate, phosphate, and oils; detecting the chemical and physical properties of samples, such as pH value, water temperature, and conductivity; and detecting the distribution of microbial communities in samples. The risk of biological water toxicity caused by microorganisms is the greatest. Especially in drinking water, the toxicity of microbial water is far greater than that of heavy metals, disinfection byproducts, and organic matter pollution. The distribution of microbial communities is becoming an increasingly important part of the indicator system. When harmful chemical elements and microorganisms exceed certain standards, corresponding safety alarms should be issued. Traditional water quality monitoring mainly relies on laboratory testing after sampling, which has limitations in terms of time and cost. In terms of direct drinking water sterilization, the methods of periodic sterilization or manual control based on user needs are clearly not intelligent enough. Summary of the Invention

[0004] To address the issues of multiple water intake types and intelligent sterilization of piped drinking water terminals, this invention provides an intelligent water intake device for drinking water terminals.

[0005] The present invention provides an intelligent water dispensing device for a direct drinking water terminal, comprising a sampling pool, a Fourier transform illumination source, an optical sensor, a rapid heating unit, a cooling unit, a water dispensing unit, a human-computer interaction unit, and a control unit;

[0006] The supply and return water pipelines are equipped with a sampling tank, a sterilization unit, a rapid heating unit, a cooling unit, and a water intake unit;

[0007] The human-machine interface unit is used to input hot water commands, cold water commands, warm water commands, quantitative water dispensing commands, and on-demand water dispensing commands, and sends the input commands to the control unit.

[0008] When the control unit receives hot water and warm water commands, it controls the rapid heating unit to work and heat the water to the set temperature. When it receives cold water commands, it controls the cooling unit to work and cool the water to the set cold water temperature. When it receives a quantitative water dispensing command, it controls the water dispensing unit to work and outputs a set amount of water. When it receives an on-demand water dispensing command, the water dispensing unit outputs water until the on-demand water dispensing command ends.

[0009] The two-dimensional array light emitted by the Fourier stacked illumination source is incident on the sampling cell at different times. The optical sensor is used to receive the diffracted light generated in the sampling cell at different times, obtain low spatial resolution three-dimensional images at each time, and send them to the control unit.

[0010] The control unit is also used to control the sampling pool to sample the water in the supply and return water pipelines. After sampling, it controls the Fourier layered illumination source and optical sensor to work, and reconstructs the low spatial resolution three-dimensional images received at various times to obtain high spatial resolution images. Based on the obtained high spatial resolution images, it determines whether sterilization is required and feeds back the sterilization signal to the upper control module, so that the upper control module controls the front-end central supply and return water device to carry out overall disinfection of the entire supply and return water system.

[0011] Preferably, the device further includes a standard solution unit and a waste discharge and cleaning unit;

[0012] The sampling pool is equipped with two inlets and one outlet, each with a valve. The first inlet of the sampling pool is connected to the water supply and return pipeline, and the second inlet of the sampling pool is connected to the standard solution unit.

[0013] The control unit is also used to control the valves of the two inlets to close and the valve of the drain outlet to open. When the water in the sampling pool is drained, the valve of the drain outlet is closed and the valve of the second inlet is opened. The standard solution in the standard solution unit is pumped into the sampling pool. At this time, the Fourier layer illumination source and optical sensor are controlled to work. The reconstructed high spatial resolution image and the standard image are compared to determine whether waste discharge and pipeline cleaning are required. If so, the waste discharge and cleaning unit is controlled to discharge waste and clean the supply and return water pipelines.

[0014] The standard image is an image obtained by sampling the standard solution in the sampling pool after waste discharge and cleaning, using a Fourier transform illumination source and an optical sensor.

[0015] Preferably, the control unit stores a detection model, which is a model based on a convolutional neural network and trained using a training set. The training set consists of high spatial resolution images of sampling pools containing drinking water in different states, including those requiring sterilization and those not requiring sterilization.

[0016] Preferably, the detection model is implemented using a convolutional neural network based on dynamic convolution, including 5 Conv3*3 convolutional layers, 2 downsampling layers, 2 upsampling layers, 2 concatenation layers, 2 serpentine convolutional layers and 1 classification layer;

[0017] The preprocessed high spatial resolution image is used as the input image and fed into Conv3*3 convolutional layer 1. The output of Conv3*3 convolutional layer 1 is activated using the ReLU activation function and then fed into downsampling layer 1 and serpentine convolutional layer 1. The output of downsampling layer 1 is fed into Conv3*3 convolutional layer 2. The output of Conv3*3 convolutional layer 2 is activated using the ReLU activation function and then fed into downsampling layer 2 and serpentine convolutional layer 2.

[0018] The output of the downsampling layer 2 enters the Conv3*3 convolutional layer 3. The output of the Conv3*3 convolutional layer 3 is activated by the ReLU activation function and then enters the upsampling layer 1. The output of the upsampling layer 1 and the output of the serpentine convolutional layer 2 enter the concatenation layer 1. The output of the concatenation layer 1 enters the Conv3*3 convolutional layer 4. The output of the Conv3*3 convolutional layer 4 is activated by the ReLU activation function and then enters the upsampling layer 2. The output of the upsampling layer 2 and the output of the serpentine convolutional layer 1 enter the concatenation layer 2. The output of the concatenation layer 2 enters the Conv3*3 convolutional layer 5. The output of the Conv3*3 convolutional layer 5 is activated by the ReLU activation function and then enters the classification layer. The output of the classification layer is then activated by the sigmoid activation function and then output.

[0019] Preferably, the device also includes a sterilization unit that uses ultraviolet light for terminal disinfection.

[0020] Preferably, the device further includes a metering unit for measuring the water flowing through the pipeline and sending the measurement data to the control unit.

[0021] Preferably, the device also includes a prepayment unit, whereby the control unit calculates water consumption based on the metering unit, displays water fees and payment methods, and reminds users to pay.

[0022] Preferably, the device further includes a data transmission module, and the control unit is used to transmit water usage information to other terminals through the data transmission module.

[0023] The beneficial effects of this invention are that it fuses light field and Fourier transform techniques to reconstruct a three-dimensional image of drinking water. This image is then used to monitor whether the drinking water in the supply and return pipes requires sterilization, thereby ensuring the quality of the water intake. This invention's water intake device can achieve automatic monitoring and sterilization control, guaranteeing the quality of drinking water. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the intelligent water dispensing device for direct drinking water terminals of the present invention. 1-Inlet valve, 2-Metering module, 3-Disinfection module, 4-Rapid heating module, 5-Cooling module, 6-Water dispensing module, 7-Waste discharge module, 8-Human-machine interaction module, 9-Data transmission module;

[0025] Figure 2 This is a schematic diagram illustrating the principle of the detection model of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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.

[0027] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0028] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0029] The intelligent water dispensing device for direct drinking water terminals in this embodiment includes a sampling pool, a Fourier transform illumination source, an optical sensor, a sterilization unit, a rapid heating unit, a cooling unit, a water dispensing unit, a human-computer interaction unit, and a control unit.

[0030] The supply and return water pipelines are equipped with a sampling tank, a sterilization unit, a rapid heating unit, a cooling unit, and a water intake unit;

[0031] The human-machine interface unit is used to input hot water commands, cold water commands, warm water commands, quantitative water dispensing commands, and on-demand water dispensing commands, and sends the input commands to the control unit.

[0032] The control unit, upon receiving hot water or warm water commands, controls the rapid heating unit to heat the water to the set temperature; upon receiving a cold water command, it controls the cooling unit to cool the water to the set cold water temperature; upon receiving a quantitative water dispensing command, it controls the water dispensing unit to output the set amount of water; and upon receiving an on-demand water dispensing command, the water dispensing unit outputs water until the on-demand water dispensing command ends. This implementation can measure drinking water volume, dispensing water on demand or in fixed quantities, and can provide multiple modes of water dispensing according to individual needs, including hot water, warm water, and ice water, available immediately without waiting for heating or cooling.

[0033] In this embodiment, the two-dimensional array light emitted by the Fourier stacked illumination source is incident on the sampling cell at staggered times. An optical sensor receives the diffracted light generated at different times in the sampling cell, obtaining low spatial resolution three-dimensional images at each moment, which are then sent to the control unit. Fourier stacked imaging means imaging by folding multiple diffraction images. In current stacked imaging, an image is obtained by illuminating different positions of the sample under test with an illumination probe; this constitutes one layer. By moving the sample to different positions, many layers can be obtained. Stacking these layers helps to lock the phase relationship at different positions of the sample. Then, a phase retrieval algorithm can be used to reconstruct this phase information.

[0034] The control unit is also used to control the sampling pool to sample water in the supply and return water pipelines. After sampling, it controls the Fourier layered illumination source and optical sensor to work, and reconstructs the low spatial resolution three-dimensional images received at various times to obtain high spatial resolution images. Based on the obtained high spatial resolution images, it determines whether sterilization is required. The images can be analyzed to determine whether the bacterial content in the images exceeds the standard. The analysis results are fed back to the upper control module to control the pre-supply and return water device to disinfect the entire supply and return water system as a whole, ensuring water safety from the source.

[0035] Fourier layered reconstruction methods utilize intensity and overlap constraints from the original image to reconstruct the object's spectrum. The basic principle of Fourier layered reconstruction is to iteratively update the original image to obtain a synthesized spectrum, resulting in the reconstructed image. Convolutional neural networks can also be used for reconstruction.

[0036] Because the sampling tank and water supply and return pipes will accumulate scale after long-term use, which will affect the water quality and the acquisition of images in the sampling tank, it is necessary to perform waste discharge cleaning on the water supply and return pipes and the sampling tank. This embodiment also includes a standard liquid unit and a waste discharge cleaning unit.

[0037] The sampling pool is equipped with two inlets and one outlet, all of which are fitted with valves. The first inlet of the sampling pool is connected to the supply and return water pipeline, and the second inlet of the sampling pool is connected to the standard solution unit.

[0038] The process of waste discharge and cleaning control by the control unit includes two stages: a detection stage and a waste discharge and cleaning stage.

[0039] During the detection phase: The control unit controls the valves of the two inlets to close and the valve of the drain outlet to open to prevent contamination of the supply and return water pipes. After the water in the sampling pool is drained, the drain outlet valve is closed and the valve of the second inlet is opened. The standard solution in the standard solution unit is pumped into the sampling pool. At this time, the Fourier layer illumination source and optical sensor are controlled to work. The reconstructed high spatial resolution image and the standard image are compared to determine whether waste discharge and pipeline cleaning are required. If so, the waste discharge and cleaning unit is controlled to discharge waste and clean the supply and return water pipes.

[0040] The standard image is an image obtained by using a Fourier transform illumination source and an optical sensor after the sampling pool has undergone waste discharge and cleaning, and after sampling the standard solution. It is used to determine whether the sampling pool needs to be cleaned and discharged.

[0041] Wastewater Discharge and Cleaning Phase: This phase can be coordinated with the central water supply and return module for pipeline circulation and chemical cleaning. After the set operating time, to ensure water supply quality, the pipeline will undergo circulation and chemical cleaning. During this cleaning process, a warning will be displayed indicating that water cannot be drawn from the equipment. After cleaning, the pipeline can be intelligently emptied and rinsed with clean water. After rinsing, the outlet valves of the supply and return water pipes will open to discharge wastewater from the branch pipes.

[0042] The control unit in this embodiment stores a detection model, which is a model based on a convolutional neural network and trained using a training set. The training set consists of high spatial resolution images of sampling pools containing drinking water in different states, including those requiring sterilization and those not requiring sterilization. The detection model provided in this embodiment is implemented using a convolutional neural network based on dynamic convolution, including 5 Conv3*3 convolutional layers, 2 downsampling layers, 2 upsampling layers, 2 stitching layers, 2 serpentine convolutional layers, and 1 classification layer.

[0043] The preprocessed high spatial resolution image is used as the input image and fed into Conv3*3 convolutional layer 1. The output of Conv3*3 convolutional layer 1 is activated using the ReLU activation function and then fed into downsampling layer 1 and serpentine convolutional layer 1. The output of downsampling layer 1 is fed into Conv3*3 convolutional layer 2. The output of Conv3*3 convolutional layer 2 is activated using the ReLU activation function and then fed into downsampling layer 2 and serpentine convolutional layer 2.

[0044] The output of the downsampling layer 2 enters the Conv3*3 convolutional layer 3. The output of the Conv3*3 convolutional layer 3 is activated by the ReLU activation function and then enters the upsampling layer 1. The output of the upsampling layer 1 and the output of the serpentine convolutional layer 2 enter the concatenation layer 1. The output of the concatenation layer 1 enters the Conv3*3 convolutional layer 4. The output of the Conv3*3 convolutional layer 4 is activated by the ReLU activation function and then enters the upsampling layer 2. The output of the upsampling layer 2 and the output of the serpentine convolutional layer 1 enter the concatenation layer 2. The output of the concatenation layer 2 enters the Conv3*3 convolutional layer 5. The output of the Conv3*3 convolutional layer 5 is activated by the ReLU activation function and then enters the classification layer. The output of the classification layer is then activated by the sigmoid activation function and then output.

[0045] While the downsampling layer in this embodiment can reduce computation and increase the receptive field, it also sacrifices some image information. When an upsampling layer is then used to restore the image size, this results in a loss of image detail. The presence of both downsampling and upsampling layers in this embodiment leads to significant loss of detail, resulting in poor segmentation performance in the convolutional neural network. Therefore, dynamic serpentine convolution is introduced to compensate for the lost detail and improve the network's receptive field.

[0046] The sterilization unit in this embodiment utilizes ultraviolet light to provide terminal disinfection at the water intake point. The ultraviolet sterilizer is a crucial component of the device and employs high-efficiency ultraviolet lamps. The ultraviolet sterilizer emits ultraviolet light of a specific wavelength, possessing strong sterilization capabilities.

[0047] The device in this embodiment also includes a metering unit and a prepaid unit. The metering unit is used to measure the water flowing through the pipeline and send the data to the control unit.

[0048] The control unit calculates water consumption based on the metering unit and displays water fees and payment methods. It is also equipped with an electric inlet valve that promptly reminds users to pay when they are in arrears. If payment is not made within the specified time, the system automatically closes the inlet valve to stop water supply. Once the user has paid the water fee, the system automatically reopens the valve, allowing the user to continue using water.

[0049] The device in this embodiment also includes a data transmission module, and the control unit is used to collect and transmit water usage information and operation data to other terminals in real time through the data transmission unit.

[0050] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A smart water dispensing device for direct drinking water terminals, characterized in that, It includes a sampling cell, a Fourier layered illumination source, an optical sensor, a rapid heating unit, a cooling unit, a water intake unit, a human-machine interaction unit, and a control unit; The supply and return water pipelines are equipped with a sampling tank, a sterilization unit, a rapid heating unit, a cooling unit, and a water intake unit; The human-machine interface unit is used to input hot water commands, cold water commands, warm water commands, quantitative water dispensing commands, and on-demand water dispensing commands, and sends the input commands to the control unit. When the control unit receives hot water and warm water commands, it controls the rapid heating unit to work and heat the water to the set temperature. When it receives cold water commands, it controls the cooling unit to work and cool the water to the set cold water temperature. When it receives a quantitative water dispensing command, it controls the water dispensing unit to work and outputs a set amount of water. When it receives an on-demand water dispensing command, the water dispensing unit outputs water until the on-demand water dispensing command ends. The two-dimensional array light emitted by the Fourier stacked illumination source is incident on the sampling cell at different times. The optical sensor is used to receive the diffracted light generated in the sampling cell at different times, obtain low spatial resolution three-dimensional images at each time, and send them to the control unit. The control unit is also used to control the sampling pool to sample the water in the supply and return water pipelines. After sampling, it controls the Fourier stacked illumination source and optical sensor to work, and reconstructs the low spatial resolution three-dimensional images received at various times to obtain high spatial resolution images. Based on the obtained high spatial resolution images, it determines whether sterilization is required and feeds back the sterilization signal to the upper control module, so that the upper control module controls the front central supply and return water device to carry out overall disinfection of the entire supply and return water system. The device also includes a standard solution unit and a waste discharge and cleaning unit; The sampling pool is equipped with two inlets and one outlet, each with a valve. The first inlet of the sampling pool is connected to the water supply and return pipeline, and the second inlet of the sampling pool is connected to the standard solution unit. The control unit is also used to control the valves of the two inlets to close and the valve of the drain outlet to open. When the water in the sampling pool is drained, the valve of the drain outlet is closed and the valve of the second inlet is opened. The standard solution in the standard solution unit is pumped into the sampling pool. At this time, the Fourier layer illumination source and optical sensor are controlled to work. The reconstructed high spatial resolution image and the standard image are compared to determine whether waste discharge and pipeline cleaning are required. If so, the waste discharge and cleaning unit is controlled to discharge waste and clean the supply and return water pipelines. The standard image is an image obtained by sampling the standard solution in the sampling pool after waste discharge and cleaning, using a Fourier transform illumination source and an optical sensor.

2. The intelligent water dispensing device for direct drinking water terminals according to claim 1, characterized in that, The control unit stores a detection model, which is a model based on a convolutional neural network and trained using a training set. The training set consists of high spatial resolution images of sampling pools containing drinking water in different states, including those requiring sterilization and those not requiring sterilization.

3. The intelligent water dispensing device for direct drinking water terminals according to claim 2, characterized in that, The detection model is implemented using a convolutional neural network based on dynamic convolution, including 5 Conv3*3 convolutional layers, 2 downsampling layers, 2 upsampling layers, 2 concatenation layers, 2 serpentine convolutional layers and 1 classification layer; The preprocessed high spatial resolution image is used as the input image and fed into the first Conv3*3 convolutional layer. The output of the first Conv3*3 convolutional layer is activated by the ReLU activation function and then fed into the first downsampling layer and the first serpentine convolutional layer. The output of the first downsampling layer is fed into the second Conv3*3 convolutional layer. The output of the second Conv3*3 convolutional layer is activated by the ReLU activation function and then fed into the second downsampling layer and the second serpentine convolutional layer. The output of the downsampling layer 2 enters the Conv3*3 convolutional layer 3. The output of the Conv3*3 convolutional layer 3 is activated by the ReLU activation function and then enters the upsampling layer 1. The output of the upsampling layer 1 and the output of the serpentine convolutional layer 2 enter the concatenation layer 1. The output of the concatenation layer 1 enters the Conv3*3 convolutional layer 4. The output of the Conv3*3 convolutional layer 4 is activated by the ReLU activation function and then enters the upsampling layer 2. The output of the upsampling layer 2 and the output of the serpentine convolutional layer 1 enter the concatenation layer 2. The output of the concatenation layer 2 enters the Conv3*3 convolutional layer 5. The output of the Conv3*3 convolutional layer 5 is activated by the ReLU activation function and then enters the classification layer. The output of the classification layer is then activated by the sigmoid activation function and then output.

4. The intelligent water dispensing device for direct drinking water terminals according to claim 1, characterized in that, The device also includes a sterilization unit that uses ultraviolet light for terminal disinfection.

5. The intelligent water dispensing device for direct drinking water terminals according to claim 1, characterized in that, The device also includes a metering unit, which measures the water flowing through the pipeline and sends the measurement data to the control unit.

6. The intelligent water dispensing device for direct drinking water terminals according to claim 1, characterized in that, The device also includes a prepayment unit. The control unit calculates water consumption based on the metering unit, displays water fees and payment methods, and reminds users to pay.

7. The intelligent water dispensing device for direct drinking water terminals according to claim 1, characterized in that, The device also includes a data transmission module, and the control unit is used to transmit water usage information to other terminals through the data transmission module.

Images