Intelligent prediction and distribution system for mountain rural water supply

By designing an intelligent prediction and distribution system in the rural water supply system in the mountainous area, using the linkage between the main reservoir and the auxiliary reservoir, combined with PLC control and prediction model, the problem of uneven flow in the village is solved, and automated monitoring and energy conservation and consumption reduction are achieved. It ensures drinking water supply in villages with high altitudes.

CN119041530APending Publication Date: 2024-11-29TIANJIN WATER CONSERVANCY SURVEYING & DESIGNING INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411379538.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-30
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

Due to the complex terrain and scattered population distribution in the rural water supply system in the mountainous areas, the flow of the pipeline network in the village is uneven, making it difficult to ensure the drinking water flow of residents in the village at a high altitude, and increasing the output frequency of the water source well pump will cause excessive losses of electricity and water resources.

Method used

Design an intelligent prediction and distribution system for rural water supply in mountainous areas, using the linkage between the main reservoir and the auxiliary reservoir, automatically adjust the water supply flow through the PLC controller and the electric regulating valve, combine the prediction model to predict water supply demand based on meteorological and historical data, and give priority to using Water collection wells and water source wells realize automated monitoring and flow coordination.

Benefits of technology

It realizes the automated monitoring and operation of rural water supply systems in mountainous areas, saves water resources and electricity, ensures drinking water quality, solves the problem of uneven flow in the village, and improves the economic and popularization of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119041530A_ABST
    Figure CN119041530A_ABST
Patent Text Reader

Abstract

The invention discloses a mountainous area rural water supply intelligent prediction distribution system, which comprises a main reservoir and a plurality of auxiliary reservoirs, and is characterized in that the main reservoir is a reservoir which is located in an administrative village, is closest to a water source and has the maximum storage capacity, and is marked as a 1 # reservoir; the small reservoirs in the natural villages affiliated to the administrative villages are marked as affiliated reservoirs; a water outlet of the main reservoir is correspondingly communicated with water inlets of the plurality of auxiliary reservoirs through a plurality of water inlet pipelines which are connected in parallel, and each water inlet pipeline is additionally provided with an electric control valve; and a water inlet of the main reservoir is connected with a water source through a water supply pipeline. Flow prediction is realized by using a recurrent neural network model optimized by a particle swarm algorithm, system building is completed by locally utilizing a high-quality water source in a mountainous area, an existing water source well and a reservoir, finally, automatic feeding and disinfection are completed from the water source, flow coordinated distribution is performed, and water supply flow among different villages is automatically adjusted and distributed. The drinking water quality is ensured while a large amount of water resources and electric energy are saved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of intelligent prediction and distribution of water supply, and in particular to an intelligent prediction and distribution system for water supply in mountainous rural areas. Background Art

[0002] Water is the source of life, and drinking water is closely related to people's life and health. Due to the special environment of mountainous areas, the terrain lines are obviously undulating, and different villages are built on the mountains with large differences in altitude. It is easy for hydraulic imbalances in different villages to cause uneven flow in the village pipeline network, and it is difficult to guarantee the drinking water flow of residents in villages downstream of the water supply pipeline or at higher altitudes. At the same time, in order to maintain the drinking water flow supply of more distant villages, the output frequency of water well pumps is increased, which can easily cause the problem of overall excessive loss of electricity and water resources.

[0003] The rural population in mountainous areas has been constantly migrating, and the population distribution has shown a trend of fragmentation. Several adjacent natural villages constitute a large administrative village, and the drinking water supply of all villagers in the natural villages must be guaranteed. The residents of villages far away from the water source or at high altitudes are scattered, which is prone to problems such as small diameter of the designed water supply network and unreasonable routing, making it difficult to ensure a continuous and sufficient flow supply. Now that my country's national economy has been continuously developed, people have gradually paid attention to the treatment of drinking water in mountainous areas, the quality assurance and improvement of rural drinking water, and have higher standards for drinking water projects. At the same time, in order to solve the problem of uneven flow in the village network, the flow of the network must be controlled within a reasonable range, which can not only meet the demand for drinking water flow in distant villages, but also make the entire water supply system more energy-saving and consumption-reducing. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings and defects of the prior art and to provide an intelligent prediction and distribution system for water supply in mountainous rural areas, which can utilize high-quality water sources in mountainous areas and existing water wells and reservoirs on the spot, and regulate them based on administrative villages in mountainous areas, and has good economy and high efficiency.

[0005] An intelligent prediction and distribution system for water supply in mountainous rural areas comprises a main water reservoir and a plurality of subsidiary water reservoirs, wherein the main water reservoir is a water reservoir located in an administrative village, which is closest to a water source and has the largest storage capacity, and is recorded as 1# water reservoir; the small water reservoirs in natural villages affiliated to the administrative village are recorded as subsidiary water reservoirs; the water outlet of the main water reservoir is correspondingly connected to the water inlet of a plurality of subsidiary water reservoirs through a plurality of parallel water inlet pipes, and each of the water inlet pipes is equipped with an electric regulating valve; the water inlet of the main water reservoir is connected to the water source through a water supply pipeline.

[0006] The water source includes a water source well and a water collection well, which are located in a mountainous area and store water that meets water quality requirements. The water from the water source well and the water collection well are respectively sent into the main water reservoir through a variable frequency pump.

[0007] Among them, there are multiple water source wells and one water collection well.

[0008] Among them, the variable frequency pump has automatic mode and manual mode, and is connected to the PLC controller. When it is in automatic control mode, the PLC controller automatically opens and closes it based on the water level of the main water reservoir detected by the water level sensor, so that the water level of the main water reservoir always meets the preset water level threshold.

[0009] Among them, the water source well and the water collection well are equipped with an automatic feeding and disinfection function, and disinfectants are regularly added to the wells to ensure that the quality of drinking water meets the preset standards.

[0010] Wherein, when water is supplied automatically, the water collection well is used preferentially for water supply.

[0011] Among them, each of the affiliated water reservoirs is equipped with a corresponding PLC control box, and a water level detection sensor is arranged to detect the water level of the corresponding affiliated water reservoir. Based on the real-time data of the water level detection sensor, the opening size of the corresponding electric regulating valve is automatically controlled to automatically adjust the water supply flow between different affiliated natural villages.

[0012] Among them, the affiliated water reservoirs of each of the natural villages affiliated to the administrative village set the corresponding reference water level according to the predicted water demand of each affiliated natural village, and perform automatic water level control and adjustment based on the reference water level; in the automatic adjustment mode, the system automatically reduces the valve opening of the electric regulating valve of the affiliated water reservoir with lower terrain or upstream of the pipeline to closed, and sets the valve opening of the electric regulating valve corresponding to the affiliated water reservoir at the end or at the highest position to 100%, so as to ensure that the flow in the affiliated water reservoir at the end or at the highest position reaches the reference water level; then the system automatically and evenly adjusts, slowly opens the valve opening of the electric regulating valve of the affiliated village water reservoir with lower terrain or upstream of the pipeline, and slowly reduces the valve opening of the electric regulating valve of the affiliated water reservoir at the end or at the highest position, until dynamic balance is achieved and the water level of each affiliated water reservoir is in the corresponding reference water level interval;

[0013] The reference water level of the subsidiary water reservoir is predicted by a prediction model. The prediction model is used to input data related to the water supply flow, including outdoor average temperature, outdoor relative humidity, solar load, precipitation, dew point, historical water volume, historical electricity volume and other related data and the water supply pressure that needs to be controlled at the next moment, and finally output the flow value at a preset moment, and divide the flow value by the bottom area of ​​the corresponding subsidiary water reservoir to obtain the reference water level.

[0014] Among them, in the process of automatic water level control and adjustment, it is first compared whether the absolute value of the real-time water level fluctuation range of each subsidiary water reservoir is less than the preset water level fluctuation range ε. If the value is less than the fluctuation range ε, the valve opening of each subsidiary water reservoir remains unchanged; if the absolute value of the real-time water level fluctuation range of a subsidiary water reservoir is greater than the preset water level fluctuation range ε, the valve of its electric regulating valve enters the automatic adjustment state and adjusts the valve opening.

[0015] Among them, the water source well is linked with the main water reservoir; when the water level in the main water reservoir does not reach the reference water level, the water source well pump is started at full power with frequency conversion until the water level in the main water reservoir reaches the preset range above and below the reference water level; after the main water reservoir reaches the normal fluctuation range of the reference water level, the water pump runs automatically with normal frequency conversion; when multiple water sources are connected, the water supply from the collection well is automatically started first, and the water source from the water source well is started to supply water together after the set adjustment time has exceeded. After the main reservoir reaches the reference water level, the water source well pump is turned off, and the adjustment time is refreshed at the same time for a new round of automatic regulation.

[0016] The present invention realizes automatic monitoring and operation of the water supply system by transforming the rural water supply system in mountainous areas. The system can be built by utilizing high-quality water sources in mountainous areas and existing water wells and reservoirs. Finally, automatic feeding and disinfection and flow coordination and distribution are completed from the water source. An online prediction system is developed to automatically adjust and distribute the water supply flow between different villages, thereby saving a large amount of water resources and electricity while ensuring the quality of drinking water.

[0017] The present invention can also use the configuration software of the host computer to achieve fault alarm, report printing, dynamic monitoring management and working condition recording for the entire water supply system, so as to ensure that the operation of the pipe network is economical, safe and real-time. The control method for water supply in mountainous rural areas has certain scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the intelligent prediction and distribution system for water supply in mountainous rural areas of the present invention.

[0019] Figure 2 It is a schematic diagram of the regulation process of the intelligent prediction and distribution system for water supply in mountainous rural areas of the present invention. DETAILED DESCRIPTION

[0020] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0021] The technical terms involved in the present invention are explained as follows:

[0022] Water source well: The water source well in this article is a deep water well, with a supporting well house, including water supply equipment and disinfection equipment, etc. A deep well water pump is installed in each well.

[0023] Water collection well: generally adopts underground rectangular structure, the pool wall structure is made of mortar masonry material, the water inlet hole is set on the water-facing side, and the graded crushed stone is wrapped with geotextile to filter the groundwater. The top cover of the pool is a reinforced concrete slab, and the well cover is a cement steel fiber well cover.

[0024] Reservoir: A reservoir is a water storage facility built with artificial materials and has an anti-seepage function.

[0025] Uniform control: Uniform control system is a control system based on adaptive control theory. It combines multiple controllers in adaptive control theory and makes adaptive selections based on the performance of the control system, thereby achieving stable control of the target variable in a changing environment. Compared with traditional control systems, uniform control systems use a parallel structure of multiple controllers, which not only ensures the robustness of the system, but also can adaptively select the optimal controller based on the control effect, thereby improving the efficiency and stability of the control system.

[0026] PLC: Programmable Logic Controller (PLC), a digital computing controller with a microprocessor for automated control, which can load control instructions into memory at any time for storage and execution. The programmable controller consists of CPU, instruction and data memory, input / output interface, power supply, digital analog conversion and other functional units.

[0027] See also Figure 1 As shown, in an embodiment of the present invention, an intelligent prediction and distribution system for water supply in mountainous rural areas includes a main water reservoir (i.e., 1# water reservoir) and multiple subsidiary water reservoirs, wherein the main water reservoir is a water reservoir located in an administrative village, which is closest to a water source and has the largest storage capacity, and is recorded as 1# water reservoir; the small water reservoirs in the natural villages affiliated to the administrative village are recorded as subsidiary water reservoirs; the water outlet of the main water reservoir is correspondingly connected to the water inlet of the multiple subsidiary water reservoirs through multiple parallel water inlet pipes, and each of the water inlet pipes is equipped with an electric regulating valve; the water inlet of the main water reservoir is connected to the water source through a water supply pipeline.

[0028] Furthermore, in some embodiments, a remote water supply control center is provided. Specifically, a remote water supply control center is provided in an administrative village, and has a built-in online prediction system. It runs a flow prediction model based on water supply pressure control, which can predict the required flow of the subsidiary water reservoir at future times and calculate the reference water level of the subsidiary water reservoir to facilitate automatic adjustment of the automatic water inlet control of the subsidiary water reservoir. In some implementation schemes, the remote water supply control center adopts the configuration software of the host computer to send the water level prediction value of the subsidiary water reservoir to the control system of the subsidiary water reservoir and perform fault alarm, report printing, dynamic monitoring management and working condition recording for the entire system, so as to ensure that the operation of the pipeline network is economical, safe and real-time.

[0029] In some embodiments, the water source includes a water source well and a water collection well, which are located in a mountainous area and store water that meets the water quality requirements. The water in the water source well and the water collection well are respectively sent to the main water reservoir through a variable frequency pump. In some embodiments, there are multiple water source wells and one water collection well. In some embodiments, the water source well and the water collection well are equipped with an automatic feeding and disinfection function, and disinfectants are regularly added to the water wells to ensure that the quality of drinking water meets the preset standards. In some embodiments, when the water is automatically supplied, the water collection well is used for water supply first.

[0030] In some embodiments, the variable frequency pump has an automatic mode and a manual mode, and is connected to a PLC controller. When in the automatic control mode, the PLC controller automatically opens and closes the pump based on the water level of the main water reservoir detected by the water level sensor, so that the water level of the main water reservoir always meets the preset water level threshold.

[0031] In some embodiments, each of the subsidiary water reservoirs is configured with a corresponding PLC control box, and a water level detection sensor is arranged to detect the water level of the corresponding subsidiary water reservoir. Based on the real-time data of the water level detection sensor, the opening size of the corresponding electric regulating valve is automatically controlled to automatically adjust the water supply flow between different subsidiary natural villages.

[0032] In some embodiments, the affiliated water reservoirs of each of the natural villages affiliated to the administrative village set corresponding reference water levels according to the predicted water demand of each affiliated natural village, and automatically control and adjust the water level based on the reference water level; in the automatic adjustment mode, the system automatically reduces the valve opening of the electric regulating valve of the affiliated water reservoirs with lower terrain or upstream of the pipeline to closed, and sets the valve opening of the electric regulating valve corresponding to the affiliated water reservoir at the end or at the highest position to 100%, ensuring that the flow in the affiliated water reservoir at the end or at the highest position reaches the reference water level; then the system automatically adjusts evenly. Slowly open the valve opening of the electric regulating valve of the affiliated village water reservoir with lower terrain or upstream of the pipeline, and slowly reduce the valve opening of the electric regulating valve of the affiliated water reservoir at the end or at the highest position until dynamic balance is achieved and the water level of each affiliated water reservoir is in the corresponding reference water level interval;

[0033] In some embodiments, the reference water level of the subsidiary water reservoir is predicted by a prediction model, which is used to input data related to the water supply flow, including outdoor average temperature, outdoor relative humidity, solar load, precipitation, dew point, historical water volume, historical electricity volume and other related data and the water supply pressure head that needs to be controlled at the next moment, and finally output the flow value at a preset moment, and the reference water level is obtained by dividing the flow value by the bottom area of ​​the corresponding subsidiary water reservoir.

[0034] In some embodiments, during the automatic water level control and adjustment process, a comparison is first made as to whether the absolute value of the real-time water level fluctuation range of each subsidiary water reservoir is less than the preset water level fluctuation range ε. If the value is less than the fluctuation range ε, the valve opening of each subsidiary water reservoir remains unchanged; if the absolute value of the real-time water level fluctuation range of a subsidiary water reservoir is greater than the preset water level fluctuation range ε, the valve of its electric regulating valve enters an automatic adjustment state to adjust the valve opening.

[0035] Among them, the water source well is linked with the main water reservoir; when the water level in the main water reservoir does not reach the reference water level, the water source well pump is started at full power with frequency conversion until the water level in the main water reservoir reaches the preset range above and below the reference water level; after the main water reservoir reaches the normal fluctuation range of the reference water level, the water pump runs automatically with normal frequency conversion; when multiple water sources are connected, the water supply from the collection well is automatically started first, and the water source from the water source well is started to supply water together after the set adjustment time has exceeded. After the main reservoir reaches the reference water level, the water source well pump is turned off, and the adjustment time is refreshed at the same time for a new round of automatic regulation.

[0036] In this application, the prediction model can be implemented in the following ways:

[0037] First, crawler technology is used to automatically crawl data from meteorological websites, collect historical water and electricity consumption data of relevant affiliated natural villages, and use NB-IoT protocol water meters to upload pipeline flow, pressure and other parameters in real time and summarize them into the database; second, the relevant parameters collected by each underlying sensor device in real time are uploaded to the SCADA software through PLC for recording, and the data is automatically uploaded to the database; third, programs are written in PyCharm to complete data query, model loading, model calculation, data upload and distribution and other functions.

[0038] Among them, the meteorological data crawled from the meteorological website data, including outdoor average temperature, outdoor relative humidity, solar load, precipitation, dew point, etc., are mainly used to accurately obtain local short-term temperature changes and precipitation changes. Since water consumption is closely related to temperature changes, when the temperature is high during the dog days, the residents' daily water consumption increases significantly, showing an upward trend; when the temperature is low, the residents' lifestyle changes accordingly, and the water consumption decreases accordingly. Precipitation has a greater impact on the water consumption of the vegetable garden and livestock feeding in the family yard. Therefore, in this application, the above-mentioned meteorological data is collected to determine the residents' water consumption.

[0039] About the collection of database data: Through crawler technology, meteorological data is obtained from meteorological websites and equipment websites according to specified rules (i.e., water meters using NB-IoT protocol upload pipeline flow, pressure and other data in real time), and the current permanent population of affiliated natural villages, the number of mechanically increased population within 15 years, the natural population growth rate, the electricity consumption of each affiliated natural village, the terrain elevation, the specifications of the reservoir and other data are collected to realize the collection of input data for the prediction model; since the electricity consumption of residents is positively correlated with the water consumption, the electricity consumption can largely reflect the frequency of residents' laundry / bathing, the number of washing machines / water dispensers / water heaters, etc., so in this application, when collecting data, the electricity consumption of each affiliated natural village should be collected to accurately determine the water consumption. Then, the specific information in the entire webpage is obtained and downloaded to the local after data analysis to realize the collection of input data for the prediction model. The connection with the database is established through the module, and the data that needs to be uploaded to the database consists of three parts: the time of obtaining the data, the name of the data and the specific value, which is convenient for the acquisition of the prediction program; the interval of crawler crawling data is set to 1 minute.

[0040] In this application, the prediction model is a prediction model that can realize the control of water supply pressure from the perspective of water demand. The input of the prediction model consists of two parts: the first part inputs historical data, that is, flow-related factors at different times, including outdoor average temperature, outdoor relative humidity, solar load, precipitation, dew point, historical water volume, historical electricity and other related data; the second part inputs the future water supply pressure control part, that is, the water supply pressure that needs to be controlled at the next moment. The water supply pressure can be calculated based on the local population, the altitude of the terrain, the volume of the affiliated reservoir and other data to determine the water supply pressure required by the local residents in the future.

[0041] In the present application, the prediction model is an online prediction model, which includes a PyCharm platform as the basis, a Keras deep learning framework training model, a trained GRU neural network model saved by the model.save() command, and a single-layer neuron; the GRU neural network model is used to process the input related data of the past moment; the single-layer neuron is used to process the input value of the water supply pressure that needs to be controlled in the future.

[0042] In the GRU neural network model layer, historical data is input into the model as two-dimensional data. The first dimension of the data is the time expansion step, and the second dimension is the flow-related factor characteristics that need to be input in one time step. Among them, the Dropout layer is set to randomly freeze a part of the neurons in the model during the training process, and remain unchanged in this iteration, and no parameter update is performed, which can effectively prevent the model from overfitting. After passing through two GRU and Dropout layers, the data will be input into the Flatten layer. The function of the Flatten layer is that the two-dimensional data needs to be flattened into one-dimensional data to facilitate the final output; finally, the historical data and the water supply pressure data at the next moment are vectorized through the Concatenate layer, and finally the flow value at the next moment is output through the fully connected layer.

[0043] Among them, the particle swarm algorithm is used to optimize the GRU neural network model, and finally the prediction model of this application is obtained: first, the constructed data set is divided into a training set and a test set; then the GRU neural network is used to train the data of the training set. First, the number of layers and the number of neurons are set to fixed values, and the changes in the evaluation indicators are observed by setting different learning rates and expanding steps at different times to determine the corresponding optimal value interval. After setting the value range of the two variables using the particle swarm algorithm, the two variables are initialized and selected, and assigned to the GRU network; the GRU network calculates the mean square error of the model in the test set based on the values ​​of the two variables. According to formulas (1) and (2), the particle speed and position are calculated, and the optimal solutions of the two parameters are obtained, and finally the required prediction model is trained.

[0044] Assuming the current time is time t, the calculation formula for the speed and position of the i-th particle at time t+1 is:

[0045]

[0046] In the formula, w is the inertia weight; c1, c2 are the learning factors of the particles; r1, R2 are random numbers between 0 and 1;

[0047] —The optimal solution under the current iteration; —Global optimal solution.

[0048] In this application, when the online prediction model is running, it can be to load the trained neural network model in keras, obtain the current time through the time function in python, and convert the obtained computer time into local time. Use the current time minus the local time of the previous 12 hours at a time interval of 30 minutes. Use the obtained time as the time index, and find the outdoor average temperature, outdoor relative humidity, solar load, precipitation, dew point, historical water volume, historical electricity and other related data of the corresponding time in the database through the time index; finally, after data processing, remove abnormal data and organize them into a two-dimensional matrix formed in the order of time, and normalize the values ​​of the matrix according to the normalization formula, add 30 minutes to the system time to get the time of the prediction, and find the water supply pressure value in the training set, and normalize the water supply pressure at the prediction time according to the normalization formula. In the specific data processing, since the historical data queried in the database is one-dimensional data, the data input into the GRU model is a two-dimensional matrix composed of different time expansion steps and different features at the same time. The hstack() command and flip() command in the NumPy library are used to organize the queried data. Finally, the data is organized into a two-dimensional matrix formed in time order, and the matrix values ​​are normalized according to the normalization formula.

[0049] The above data constitutes the data set required for the prediction time. The data set is forward propagated using the GRU neural network to calculate the predicted value of the water supply flow prediction model for each affiliated natural village, and multiple predicted values ​​are denormalized to calculate the flow value at the next moment; finally, multiple predicted values ​​are passed to the database, including the current time, the name of the affiliated natural village, and the predicted flow value V i , and then determine the water level reference value H of each subsidiary reservoir according to formula (3) o .

[0050]

[0051] Where, s is the bottom area of ​​the reservoir.

[0052] Then, the SCADA software is used to query the database for the latest predicted flow value V uploaded to the database. i And water level reference value H o , the water level reference value H is obtained through the set timer o And sent to the corresponding PLC, which will calculate the water level based on the water level reference value H o The current water level data of each reservoir automatically controls the water level of the reservoir, that is, the water level fluctuation range ε ​​is preset, and the difference between the current water level of each reservoir and the reference water level is calculated to determine the water level fluctuation range ΔH i , the two are compared, and the automatic operation is controlled according to the comparison result, such as Figure 2 shown.

[0053] Furthermore, in order to realize online model prediction, you can also use the while True command to let the program execute at a certain interval. Since the traffic is predicted at intervals of 30 minutes, set the program to execute the next prediction after 1800 seconds after making a prediction.

[0054] In this application, sensors are set up to complete the collection of water level, pressure, flow, water quality data, water pump parameters and frequency converter data in the water source well / collection well, and automatic control of water quality treatment and water pump frequency conversion control for automatic addition of disinfectant are performed to ensure automatic adjustment and control of the water level in the main water reservoir. The water level, pressure and flow of the main water reservoir are used to automatically control or switch the access or disconnection of the water source to ensure that the water level in the main water reservoir can reach the reference water level and meet the requirements.

[0055] Among them, the linkage process between the water source well and the administrative village 1# reservoir is as follows; when the water level line in the 1# reservoir does not reach the reference water level, the water source well pump frequency conversion starts at full power until the water level in the 1# reservoir reaches near the reference water level; after the 1# reservoir reaches the normal fluctuation range of the reference water level, the water pump automatically operates with normal frequency conversion. At the same time, the water source well subsystem automatically controls the start of the feeding and disinfection equipment by monitoring the water quality data to ensure the normal quality of drinking water. When multiple water sources are connected, the water supply from the water collection well is automatically started first. After the set adjustment time is exceeded, the water source well water supply is started together. After the 1# reservoir reaches the reference water level, the water source well pump is turned off, and the adjustment time is refreshed for a new round of automatic regulation.

[0056] In this application, during operation, the water level in each subsidiary water reservoir and the feedback signal of the opening of the electric regulating valve installed on the access pipe network are collected; the above collected signals are transmitted to each PLC controller corresponding to the subsidiary water reservoir. The analog input / output module contained in each PLC controller performs data conversion, and the program is designed to calculate the difference in the CPU module register to determine the water level difference and flow difference, which is used for automatic control of flow regulation and automatic control of the flow of the subsidiary water reservoir. At the same time, the corresponding pipe network parameter signals collected are transmitted to the remote control center via the village network.

[0057] In the present application, the adjustment of the electric regulating valve on the water inlet pipeline is automatic adjustment, and the adjustment process is as follows: run the control instructions of each PLC controller, and preliminarily set the control parameters of each reservoir system from near to far. Automatically distribute the water supply flow between the villages in the mountainous area, and each natural village reservoir sets the corresponding reference water level according to the water demand of each village. The natural village reservoir with lower terrain or upstream of the pipeline will quickly reach the reference water level. At this time, the system automatically reduces the valve opening of the electric regulating valve installed in the corresponding reservoir to closed; ensure that the flow in the last reservoir or the reservoir with the highest terrain reaches the reference water level, and the corresponding electric regulating valve valve opening is 100% at this time; then the system automatically adjusts evenly; slowly open the valve opening of the electric regulating valve of the natural village reservoir with lower terrain or upstream of the pipeline, and slowly reduce the valve opening of the electric regulating valve of the last reservoir or the reservoir with the highest terrain until dynamic balance is achieved, and the water level of each reservoir in each natural village is in the corresponding reference water level interval.

[0058] During the adjustment process, first compare whether the absolute value of the real-time water level fluctuation range of each reservoir is less than the preset water level fluctuation range ε. If the value is less than the fluctuation range ε, the valve opening of each reservoir remains unchanged; if the absolute value of the real-time water level fluctuation range of a reservoir is greater than the preset water level fluctuation range ε, the valve of this reservoir enters the automatic adjustment state, and the corresponding reservoir valve opening needs to be adjusted.

[0059] In this application, during the process of automatic control and adjustment, the control parameters are further uniformly adjusted, and combined with the idea of ​​uniform control, the parameters of each controller corresponding to the excessive error range of the control system output during the system adjustment process are optimized. When the pipe network system is disturbed, the opening of the electric regulating valves of each branch of the pipe network system is dynamically optimized, so that the water level parameters of the front and rear branch reservoirs are coordinated and taken into account, so as to meet the dynamic hydraulic balance of the system, and finally realize the automatic, safe and stable operation of the system, so that the flow of each affiliated natural village is coordinated and stable, and the system operates safely and stably.

[0060] The present invention fully considers the characteristics of water supply pressure and influencing factors in mountainous rural areas, effectively controls the water supply pressure at the next moment through a multi-input method, and utilizes a recurrent neural network in combination with the characteristics of a time series of water supply flow sequence, effectively extracts time series information from historical moments, improves the accuracy of prediction, and can realize the control of water supply pressure at future moments, thereby improving the practical application value of the prediction model.

[0061] The present invention effectively improves the problem of uneven distribution of drinking water flow among residents in mountainous villages with large altitude differences and large differences in distance from water sources. After the system is built, it can automatically adjust and distribute the water supply flow between different villages to ensure the drinking water flow supply for residents in villages downstream of the water supply pipeline or at higher altitudes.

[0062] The water source well and the water reservoir of the present invention operate in linkage, which is different from the existing conventional water supply operation mode. At present, the water source well pump generally operates at industrial frequency for 24 hours, causing huge power loss and energy waste. The water source well pump of the present invention is in variable frequency operation state under normal working conditions, which can greatly save electricity while ensuring the quality of water supply. If the system has multiple water sources, the water source of the water collection well is used first, making the overall system more energy-saving and consumption-reducing.

[0063] The present invention can accept the incorporation of multiple water sources. The water source points of the system can reasonably utilize high-quality water sources in mountainous areas. After the water quality monitoring is passed, a water collection well can be built on site to avoid the waste of water resources and the waste of funds and environmental pollution caused by too many wells.

[0064] For the close regulation process between the subsidiary water storage tanks, if the regulating valve is frequently actuated, the branch flow interferes with each other, and the hydraulic working conditions of the pipe network change frequently, it is easy to cause system oscillation. Therefore, the present invention combines the idea of ​​uniform control to make the control action relatively moderate, optimize the parameters of each controller corresponding to the excessive error range of the control system output during the system regulation process, and when disturbed, the water level of the proximal branch produces a small fluctuation within the allowable range. At the same time, the water level of the distal branch also changes slowly and in a balanced manner, thereby better eliminating the interference effect and enhancing the adaptive ability of the pipe network system. Dynamically optimize the opening of the electric regulating valves of each branch of the pipe network system, so that the water level parameters of the front and rear unit branches are coordinated with each other and taken into account, thereby satisfying the dynamic hydraulic balance of the system.

[0065] The present invention utilizes the CPU host and expansion module of the PLC to collect the operating equipment status and parameters of water wells, water source wells, water reservoirs and other places, completes the automatic issuance of relevant instructions, and finally provides automatic control of equipment and linkage and interlocking control of related equipment. It has good reliability and can effectively save water and play a role in energy conservation and emission reduction.

[0066] The present invention applies the intelligent control system to the water supply in mountainous rural areas and optimizes the regulation mode of the drinking water system, which not only improves the operating efficiency and quality of the water source well pump room, but also solves the traditional management mode of dedicated personnel and realizes the automatic collection and measurement and control of water resource information.

[0067] The basic principles and main features of the present invention and the advantages of the present invention are shown and described above. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention.

[0068] Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the claims are included in the present invention.

[0069] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. Intelligent prediction and distribution system for water supply in mountainous rural areas, characterized by: It includes a main water reservoir and multiple subsidiary water reservoirs. The main water reservoir is the water reservoir located in an administrative village, which is closest to the water source and has the largest storage capacity, and is recorded as 1# water reservoir; the small water reservoirs in the natural villages affiliated to the administrative village are recorded as subsidiary water reservoirs; the water outlet of the main water reservoir is correspondingly connected to the water inlet of multiple subsidiary water reservoirs through multiple parallel water inlet pipes, and each of the water inlet pipes is equipped with an electric regulating valve; the water inlet of the main water reservoir is connected to the water source through a water supply pipeline.

2. According to claim 1, the intelligent prediction and distribution system for water supply in mountainous rural areas is characterized by: The water source includes a water source well and a water collection well, which are located in a mountainous area and store water that meets water quality requirements. The water from the water source well and the water collection well are respectively sent into the main water reservoir through a variable frequency pump.

3. According to claim 2, the intelligent prediction and distribution system for water supply in mountainous rural areas is characterized in that: There are multiple water source wells and one water collection well.

4. According to claim 3, the intelligent prediction and distribution system for water supply in mountainous rural areas is characterized in that: The variable frequency pump has an automatic mode and a manual mode and is connected to a PLC controller. When in the automatic control mode, the PLC controller automatically opens and closes the pump based on the water level of the main water reservoir detected by the water level sensor, so that the water level of the main water reservoir always meets the preset water level threshold.

5. According to claim 4, the intelligent prediction and distribution system for water supply in mountainous rural areas is characterized in that: The water source well and the water collection well are equipped with an automatic feeding and disinfection function, and disinfectants are regularly added to the wells to ensure that the quality of drinking water meets the preset standards.

6. The intelligent prediction and distribution system for water supply in mountainous rural areas according to claim 5 is characterized in that: When water is supplied automatically, the water collection well is used first for water supply.

7. The intelligent prediction and distribution system for water supply in mountainous rural areas according to claim 1 is characterized in that: Each of the subsidiary water reservoirs is configured with a corresponding PLC control box, and a water level detection sensor is arranged to detect the water level of the corresponding subsidiary water reservoir. Based on the real-time data of the water level detection sensor, the opening size of the corresponding electric regulating valve is automatically controlled to automatically adjust the water supply flow between different subsidiary natural villages.

8. The intelligent prediction and distribution system for water supply in mountainous rural areas according to claim 7 is characterized in that: The affiliated water reservoirs of each of the natural villages affiliated to the administrative village are set with corresponding reference water levels according to the predicted water demand of each affiliated natural village, and automatic water level control and adjustment are performed based on the reference water level; in the automatic adjustment mode, the system automatically reduces the valve opening of the electric regulating valve of the affiliated water reservoirs with lower terrain or upstream of the pipeline to closed, and sets the valve opening of the electric regulating valve corresponding to the affiliated water reservoir at the end or at the highest position to 100%, so as to ensure that the flow in the affiliated water reservoir at the end or at the highest position reaches the reference water level; then the system automatically and evenly adjusts, slowly opens the valve opening of the electric regulating valve of the affiliated village water reservoirs with lower terrain or upstream of the pipeline, and slowly reduces the valve opening of the electric regulating valve of the affiliated water reservoir at the end or at the highest position, until dynamic balance is achieved and the water level of each affiliated water reservoir is in the corresponding reference water level interval; The reference water level of the subsidiary water reservoir is predicted by a prediction model. The prediction model is used to input data related to the water supply flow, including outdoor average temperature, outdoor relative humidity, solar load, precipitation, dew point, historical water volume, historical electricity volume and other related data and the water supply pressure that needs to be controlled at the next moment, and finally output the flow value at a preset moment, and divide the flow value by the bottom area of ​​the corresponding subsidiary water reservoir to obtain the reference water level.

9. The intelligent prediction and distribution system for water supply in mountainous rural areas according to claim 8 is characterized in that: During the automatic water level control and adjustment process, first compare whether the absolute value of the real-time water level fluctuation range of each subsidiary water reservoir is less than the preset water level fluctuation range ε. If the value is less than the fluctuation range ε, the valve opening of each subsidiary water reservoir remains unchanged; if the absolute value of the real-time water level fluctuation range of a subsidiary water reservoir is greater than the preset water level fluctuation range ε, the valve of its electric regulating valve enters the automatic adjustment state and adjusts the valve opening.

10. The intelligent prediction and distribution system for water supply in mountainous rural areas according to claim 1 is characterized in that: The water source well is linked with the main water reservoir; when the water level in the main water reservoir does not reach the reference water level, the water source well pump is started at full power by frequency conversion until the water level in the main water reservoir reaches the preset range above and below the reference water level; after the main water reservoir reaches the normal fluctuation range of the reference water level, the water pump runs automatically by normal frequency conversion; when multiple water sources are connected, the water supply from the water collection well is automatically started first, and the water source well is started to supply water together after the set adjustment time has expired. After the main water reservoir reaches the reference water level, the water source well pump is shut down, and the adjustment time is refreshed at the same time for a new round of automatic regulation.