CN110284556B - Energy-saving peak shifting intelligent water supply system based on cloud technology - Google Patents

Energy-saving peak shifting intelligent water supply system based on cloud technology Download PDF

Info

Publication number
CN110284556B
CN110284556B CN201910612764.9A CN201910612764A CN110284556B CN 110284556 B CN110284556 B CN 110284556B CN 201910612764 A CN201910612764 A CN 201910612764A CN 110284556 B CN110284556 B CN 110284556B
Authority
CN
China
Prior art keywords
water
real
water storage
data
time
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910612764.9A
Other languages
Chinese (zh)
Other versions
CN110284556A (en
Inventor
邓帮武
邓卓志
郑其元
陈晔斌
李威
程建国
张孝伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Anhui Shunyu Water Affairs Co Ltd
Original Assignee
Anhui Shunyu Water Affairs Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Anhui Shunyu Water Affairs Co Ltd filed Critical Anhui Shunyu Water Affairs Co Ltd
Priority to CN201910612764.9A priority Critical patent/CN110284556B/en
Publication of CN110284556A publication Critical patent/CN110284556A/en
Application granted granted Critical
Publication of CN110284556B publication Critical patent/CN110284556B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • E—FIXED CONSTRUCTIONS
    • E03—WATER SUPPLY; SEWERAGE
    • E03B—INSTALLATIONS OR METHODS FOR OBTAINING, COLLECTING, OR DISTRIBUTING WATER
    • E03B11/00—Arrangements or adaptations of tanks for water supply
    • E—FIXED CONSTRUCTIONS
    • E03—WATER SUPPLY; SEWERAGE
    • E03B—INSTALLATIONS OR METHODS FOR OBTAINING, COLLECTING, OR DISTRIBUTING WATER
    • E03B7/00—Water main or service pipe systems
    • E03B7/07—Arrangement of devices, e.g. filters, flow controls, measuring devices, siphons, valves, in the pipe systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A20/00—Water conservation; Efficient water supply; Efficient water use
    • Y02A20/15—Leakage reduction or detection in water storage or distribution
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A20/00—Water conservation; Efficient water supply; Efficient water use
    • Y02A20/152—Water filtration

Abstract

The invention relates to the field of intelligent water supply management systems, and discloses an energy-saving peak shifting intelligent water supply system based on a cloud technology. The invention has the advantages that the water storage tank can be comprehensively regulated, stored and optimally distributed to pipe network water supply resources during non-water-consumption peak, the instant water demand to the municipal water supply network is reduced during water consumption peak, the pressure fluctuation range of the water supply network is shortened, the pressure fluctuation of the water supply network is small, the leakage phenomenon of the pipe network can be reduced, the water supply stability of a water plant system is ensured, the reconstruction requirements of the municipal water supply network and the water supply pipeline assembly of the water plant are reduced, and the national water supply resources and supporting facilities are saved.

Description

Energy-saving peak shifting intelligent water supply system based on cloud technology
Technical Field
The invention relates to the field of intelligent water supply management systems, and particularly belongs to an energy-saving peak shifting intelligent water supply system based on a cloud technology.
Background
In recent years, with the acceleration of the domestic urbanization construction process of China, domestic population is continuously gathered to large and medium-sized cities, so that the quantity of the domestic large and medium-sized cities is continuously increased. However, in the process of increasing the size of large and medium-sized cities, the state-related land policies must be observed, and the large and medium-sized cities are required to improve the effective utilization rate of the land. The building for the large and medium-sized cities in China is developed towards high-rise and small and high-rise buildings, and the high-rise and small and high-rise buildings exceed six floors. According to the relevant regulations of national city construction: when the number of floors of buildings for construction such as districts and commercial office buildings of a city is more than six floors, matched secondary water supply equipment must be installed and arranged to ensure stable water supply for the buildings.
In about 2000 years, the pipe network pressure-superposed (non-negative pressure) water supply equipment in the secondary water supply equipment is widely popularized and used in many urban constructions. However, in the actual use process, the pipe network pressure-superposed (non-negative pressure) water supply equipment has higher requirements on matched water supply infrastructure due to the lack of water storage capacity, and particularly, when the water supply capacity of a municipal water plant is insufficient, the water robbing phenomenon often occurs, so that the problem of conflict between the water plant and a water supply network is more and more serious. Such as: the city that uses the pipe network to fold pressure (no negative pressure) water supply equipment by a large scale, during city water peak period, when water supply capacity of water works can not satisfy the actual water demand, water supply pipe network pressure fluctuation is violent, robbes water phenomenon serious for pipe network leakage phenomenon takes place occasionally, and water works and water supply pipe network reconstruction demand increase.
And to the part in the city use the secondary water supply equipment with water storage box, this kind of secondary water supply equipment can hold full water storage box earlier when the peak of using water, and when the peak of using water, the water level in the water storage box descends to the take the altitude to be, and the inlet pipe valve of water storage box among this kind of secondary water supply equipment can keep full open state, continues to supply water to the water storage box, and the float valve closes the inlet pipe valve and stops intaking when the water storage box water storage capacity reaches certain liquid level. The secondary water supply equipment realizes the regulation and storage of water supply in a staggered mode through regulating and controlling the water storage tank. However, the water in the water storage tank of the secondary water supply equipment has low update rate, the safety and reliability of the water quality of the water supply cannot be guaranteed, the water supply resources of a pipe network cannot be regulated, stored and optimally distributed comprehensively, the energy-saving peak-shifting intelligent water supply cannot be realized, and the actual life needs of people cannot be met. Therefore, the invention provides an energy-saving peak shifting intelligent water supply system based on a cloud technology.
Disclosure of Invention
The invention provides an energy-saving peak-shifting intelligent water supply system based on cloud technology, which can solve the problems mentioned in the technical background, can realize the purpose of comprehensively regulating, storing and optimizing a water supply network for a water storage tank at the non-water-consumption peak time, reduce the instant water demand of a municipal water supply network at the water consumption peak time, shorten the pressure fluctuation range of the municipal water supply network and a water supply network component, reduce the pressure fluctuation of the municipal water supply network and the water supply network component, simultaneously avoid the pressure fluctuation of the municipal water supply network from increasing the instant water supply pressure of a water plant, reduce the leakage loss of the municipal water supply network and the water supply network component, ensure the stable operation of the municipal water supply network and the water supply network component of the water plant, reduce the reconstruction requirements of the municipal water supply network and the water supply network component of the water plant, and realize the saving of national water supply resources and supporting facilities.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
the energy-saving peak shifting intelligent water supply system based on the cloud technology comprises a pump room, a plurality of water storage tanks, a plurality of water pump units and water supply pipeline assemblies, wherein the water storage tanks are installed in the pump room and are connected with the water pump units through the water supply pipeline assemblies, the water storage tanks of the pump room are connected with the water pump units through the water supply pipeline assemblies, the water pump rooms are internally provided with a water storage tank, the water storage tanks are connected with water supply users of the water pump units through the water supply pipeline assemblies, and the pump rooms are connected through a municipal water supply pipeline network The water quality monitoring and processing unit, the pipe network hydraulic data monitoring unit, the pump room, the water storage tank, the water pump unit and the water supply pipeline assembly are connected with each other, the water quality monitoring and processing unit, the pipe network hydraulic data monitoring unit, the water storage tank, the water pump unit and the water supply pipeline assembly of the intelligent water tank regulation and control unit hydraulic data monitoring unit can transmit hydraulic water quality information data of the pump room, the water storage tank, the water pump unit and the water supply pipeline assembly which are collected in real time to the cloud platform server through the pump room monitoring unit and the data collection and transmission unit, and instruct the cloud platform big data analysis module according to the cloud platform server to regulate the pump room, the water storage tank, the water pump unit and the water supply pipeline assembly which exceed preset standard hydraulic water quality information data so that the hydraulic water quality information data exceeding the preset standard are regulated to meet the standard, the cloud platform big data analysis module is, The intelligent water tank water pump system comprises pump room monitoring units, a smart water tank regulating and controlling unit, a hydraulic data monitoring unit, a water quality monitoring and processing unit, a pipe network hydraulic data monitoring unit and a plurality of pump room water storage tank water pump units, wherein the pump room monitoring units of all pump rooms can be used for sending and transmitting monitoring data of the smart water tank regulating and controlling units, the hydraulic data monitoring units, the water quality monitoring and processing units and the pipe network hydraulic data monitoring units in the pump rooms, the pump room monitoring units of all pump rooms send real-time data to a cloud platform big data analysis module through data acquisition and transmission units of the pump rooms, the cloud platform big data analysis module determines target water storage amount of water storage tanks corresponding to the real-time data based on a pre-trained target neural network model, and then the intelligent water tank regulating and controlling units work state data based on the real-time water storage amount of the corresponding water storage tanks, The real-time water supply data and the hydraulic data at the tail end of the pipe network monitored by the intelligent water tank regulating and controlling unit and the pipe network hydraulic data monitoring unit determine whether to store water in the water storage tank;
the pump room monitoring unit of each pump room can be used for monitoring the pressure, flow, real-time water supply quantity and real-time water consumption data of the user, which are sent by the hydraulic data monitoring unit of the pump room, for supplying water to the user;
the pump room monitoring unit of each pump room can be used for monitoring the hydraulic data of the pressure, the flow, the real-time water supply amount and the real-time water supply amount data of the water supplied to the pump room water storage tank by the pipeline sent by the pipe network hydraulic data monitoring unit of the pump room;
the pump room monitoring unit of each pump room can be used for controlling the water quality monitoring and processing unit, the water quality analyzing equipment and the water quality sterilizing equipment of the pump room;
the pump room monitoring unit of each pump room can be used for analyzing the water quality of the water storage tank by the intelligent water tank regulating and controlling unit of the pump room, monitoring and controlling the water quality of the water storage tank in real time, transmitting the water quality data information monitored in real time to the cloud platform server and the cloud platform big data analysis module through the pump room monitoring unit and the data acquisition and transmission unit, and instructing the cloud platform big data analysis module to disinfect the water body with the water quality exceeding the preset standard according to the cloud platform server;
the cloud technology used by the cloud platform server and the cloud platform big data analysis module comprises public cloud, private cloud and mixed cloud, the cloud technology can store system data in a distributed storage mode, a redundant storage mode and a cold and hot backup storage mode, and the cloud technology enables a cloud system to construct distributed computing, utility computing, load balancing computing and parallel computing through virtualization basic hardware resources;
the energy-saving peak-shifting intelligent water supply system based on the cloud technology can be used for carrying out data communication and instruction transmission by constructing a unified platform consisting of a plurality of cloud platform servers or storage server hardware and connecting a pump room, a water storage tank, a water pump unit, a water supply pipeline assembly, a computer and matched equipment in the pump room through a local area network technology;
the energy-saving peak shifting intelligent water supply system based on the cloud technology can construct a local area network based on a public network by adopting an IPsec VPN virtual private network technology based on public communication network infrastructure, and realizes data transmission between a cloud end and matched equipment in each pump room;
the energy-saving peak shifting intelligent water supply system based on the cloud technology uses a PLC as a pump room field controller to acquire field data and receive a control instruction, and performs data communication with a cloud end through a Modbus industrial bus protocol standard;
the storage of the pump room data in the energy-saving peak-shifting intelligent water supply system based on the cloud technology is realized by a distributed database, and the file processing is realized by the processing of a distributed file system;
the energy-saving peak shifting smart water supply system based on the cloud technology acquires pump room data and completes cloud storage by constructing a local area network compatible with a public communication network at a cloud end, and a cloud platform big data analysis module is applied to a cloud platform server end to analyze water use data of each pump room so as to send a regulation and control instruction;
the cloud platform big data analysis module is used for determining the real-time water consumption of a user of the intelligent water tank regulation and control unit and the real-time data sent by the data acquisition and transmission unit of each pump room, determining the target water storage amount of the water storage tank corresponding to the real-time data based on a pre-trained target neural network model for the real-time data of each pump room, sending a starting instruction to the intelligent water tank regulation and control unit when the real-time water storage amount of the corresponding water storage tank in the real-time data does not reach the target water storage amount and the intelligent water tank regulation and control unit is not opened, so that the intelligent water tank regulation and control unit starts to store water for the water storage tank, calculating a first target water storage time based on the real-time water storage amount in the real-time data, the highest water storage amount of the water storage tank sent by the intelligent water tank regulation and control unit and the target water storage amount, and sending a first stopping instruction to the intelligent water tank, the intelligent water tank regulating and controlling unit is closed, when the real-time water storage amount in the real-time data does not reach the target water storage amount and the intelligent water tank regulating and controlling unit is opened, a second target water storage time is calculated based on the real-time water storage amount in the real-time data, the rated water supply amount of the intelligent water tank regulating and controlling unit and the target water storage amount, when the second target water storage time is reached, a second stop instruction is sent to the intelligent water tank regulating and controlling unit so that the intelligent water tank regulating and controlling unit stops storing water to the water storage tank, when the real-time water storage amount in the real-time data reaches the target water storage amount and the intelligent water tank regulating and controlling unit is opened for storing water, a third stop instruction is sent to the intelligent water tank regulating and controlling unit so that the intelligent water tank regulating and controlling unit stops storing water to the water storage tank, the target neural network model is based on real-time water consumption of a user, real-time water consumption of the user and sample water, training an initial target neural network model to obtain an initial neural network model, wherein the target neural network model can enable user real-time water consumption sample data to be associated with sample water storage amount of a corresponding water storage tank, and the user real-time water consumption sample data comprises the sample real-time water storage amount of the corresponding water storage tank, the sample real-time user water consumption, sample accumulated water supply amount, sample real-time water quality data and sample real-time water supply amount;
the cloud platform big data analysis module can be further used for performing discrete cosine transform on the real-time data based on a discrete cosine transform matrix to obtain redundancy-removed real-time data before determining the target water storage capacity of the water storage tank corresponding to the real-time data based on a pre-trained target neural network model, calculating a mean value and a standard deviation of the redundancy-removed real-time data, normalizing the redundancy-removed real-time data based on the mean value and the standard deviation to obtain normalized real-time data, and determining the target water storage capacity of the water storage tank corresponding to the normalized real-time data based on the pre-trained target neural network model;
the cloud platform big data analysis module can be specifically used for arranging real-time water storage amount, real-time user water consumption, accumulated water supply amount, real-time water quality data and sampling point data contained in the real-time water supply amount of a corresponding water storage tank in real-time data into a first matrix, calculating a product of the first matrix and a discrete cosine transform matrix to obtain a second matrix, and taking each element in the second matrix as the real-time data corresponding to the pump room after redundancy removal.
Preferably, the cloud platform big data analysis module may be further configured to calculate a first mean value and a first standard deviation of first elements representing real-time water storage capacity in the second matrix, calculate, for each first element, a first difference value between the first element and the first mean value corresponding to the first element, and calculate a first quotient of the first difference value and the first standard deviation;
calculating a second mean value and a second standard deviation of second elements representing the water consumption of real-time users in the second matrix, calculating a second difference value of each second element and the corresponding second mean value, and calculating a second quotient of the second difference value and the second standard deviation;
calculating a third mean value and a third standard deviation of third elements representing accumulated water supply in the second matrix, calculating a third difference value of each third element and the third mean value corresponding to the third element, and calculating a third quotient of the third difference value and the third standard deviation;
calculating a fourth mean value and a fourth standard deviation of a fourth element representing real-time water quality data in the second matrix, calculating a fourth difference value of each fourth element and the corresponding fourth mean value, and calculating a fourth quotient of the fourth difference value and the fourth standard deviation;
calculating a fifth mean value and a fifth standard deviation of fifth elements representing real-time water supply in the second matrix, calculating a fifth difference value of each fifth element and the corresponding fifth mean value, and calculating a fifth quotient value of the fifth difference value and the fifth standard deviation;
and taking the first quotient value, the second quotient value, the third quotient value, the fourth quotient value and the fifth quotient value as the normalized real-time data.
Preferably, the training process of the target neural network model specifically includes:
acquiring user real-time water consumption sample data used for training and sample water storage amount of a corresponding water storage tank, wherein the user real-time water consumption sample data comprises the sample real-time water storage amount of the corresponding water storage tank, the sample real-time user water consumption, the sample accumulated water supply amount, the sample real-time water quality data and the sample real-time water supply amount;
inputting the real-time water consumption sample data of the user and the sample water storage amount of the corresponding water storage tank into an initial neural network model, wherein the initial neural network model comprises an input layer, a hidden layer, an output layer and an equation set relation established between the input layer and the output layer, and the equation set relation established between the input layer and the output layer is an excitation function;
according to the user real-time water consumption sample data and the sample water storage amount of a corresponding water storage tank, the weight coefficient between each node of the input layer and each node of the hidden layer, the weight coefficient between each node of the hidden layer and each node of the output layer and the equation set relation established between the input layer and the output layer;
solving the equation set to obtain the values of the weight coefficients between each node of the input layer and each node of the hidden layer and the values of the weight coefficients between each node of the hidden layer and the nodes of the output layer;
substituting the values of the weight coefficients between the nodes of the input layer and the nodes of the hidden layer, the values of the weight coefficients between the nodes of the hidden layer and the nodes of the output layer and the user real-time water consumption sample data into the equation set to solve the initial water storage capacity of the water storage tank corresponding to the user real-time water consumption sample data;
calculating a difference value between the initial water storage amount of the corresponding water storage tank and the sample water storage amount of the corresponding water storage tank;
when the difference value is larger than a preset difference threshold value, adjusting a first weight coefficient and a second weight coefficient based on the difference value, and returning to the step of inputting the user real-time water consumption sample data and the sample water storage amount of the corresponding water storage tank into an initial neural network model, wherein the first weight coefficient is a first weight coefficient between a first input layer node and a first hidden layer node, and the second weight coefficient is a second weight coefficient between a second input layer node and the first hidden layer node;
when the difference value is not greater than a preset difference threshold value, finishing training to obtain a target neural network model containing the corresponding relation between the real-time water consumption sample data of the user and the sample water storage capacity of the corresponding water storage tank;
the input layer comprises a first input layer node, a second input layer node and a third input layer node, the hidden layer comprises a first hidden layer node, a second hidden layer node and a third hidden layer node, the output layer comprises an output layer node, and the step of establishing the equation set relationship between the input layer and the output layer according to the user real-time water consumption sample data and the sample water storage capacity of the corresponding water storage tank, the weight coefficient between each node of the input layer and each node of the hidden layer, the weight coefficient between each node of the hidden layer and the output layer node of the output layer and the equation set relationship between the input layer and the output layer comprises the following steps:
establishing a first linear equation according to the user real-time water consumption sample data, a first weight coefficient between the first input layer node and the first hidden layer node, a second weight coefficient between the second input layer node and the first hidden layer node and a first constant between the third input layer node and the first hidden layer node;
establishing a second linear equation according to the user real-time water consumption sample data, a third weight coefficient between the first input layer node and the second hidden layer node, a fourth weight coefficient between the second input layer node and the second hidden layer node, and a second constant between the third input layer node and the second hidden layer node;
establishing a third linear equation according to the user real-time water consumption sample data, a fifth weight coefficient between the first input layer node and the third hidden layer node, a sixth weight coefficient between the second input layer node and the third hidden layer node, and a third constant between the third input layer node and the third hidden layer node;
establishing a fourth function equation according to the equation set relation established between the input layer and the output layer, wherein the equation set relation is an excitation function, the sample water storage capacity of a water storage tank corresponding to the user real-time water consumption sample data, the function value of the first linear equation, the function value of the second linear equation and the function value of the third linear equation, a seventh weight coefficient between the first hidden layer node and the output layer node, an eighth weight coefficient between the second hidden layer node and the output layer node and a ninth weight coefficient between the third hidden layer node and the output layer node;
the first linear equation, the second linear equation, the third linear equation and the fourth function equation are simultaneously established to obtain an equation set between the input layer and the output layer;
the step of solving the equation set to obtain the values of the weight coefficients between the nodes of the input layer and the nodes of the hidden layer and the values of the weight coefficients between the nodes of the hidden layer and the nodes of the output layer includes:
solving the equation set to obtain a value of the first weight, a value of the second weight, a value of the third weight, a value of the fourth weight, a value of the fifth weight, a value of the sixth weight, a value of the seventh weight, a value of the eighth weight, and a value of the ninth weight;
substituting the values of the weight coefficients between the nodes of the input layer and the nodes of the hidden layer, the values of the weight coefficients between the nodes of the hidden layer and the nodes of the output layer and the user real-time water consumption sample data into the equation set to solve the initial water storage capacity of the water storage tank corresponding to the user real-time water consumption sample data, wherein the step comprises the following steps:
and substituting the value of the first weight, the value of the second weight, the value of the third weight, the value of the fourth weight, the value of the fifth weight, the value of the sixth weight, the value of the seventh weight, the value of the eighth weight, the value of the ninth weight and the user real-time water consumption sample data into the equation set to solve the initial water storage capacity of the water storage tank corresponding to the user real-time water consumption sample data.
Preferably, the function value of the excitation function is: the sum of the excitation coefficient and a target maximum value, wherein the target maximum value is the maximum value in the user real-time water consumption sample data and a target product, and the target product is the product of the excitation coefficient and the user real-time water consumption sample data.
Preferably, still install in the water storage box of every pump house and be provided with sterilizer and water storage box self-cleaning equipment, cloud platform big data analysis module can also be used for confirming whether quality of water in the water storage box of this pump house reaches preset standard according to the real-time quality of water data in this real-time data, and when quality of water does not reach preset standard for lasting long time reach first preset duration, send the sterilizer and open the sterilizer of instruction to this pump house water storage box in, so that the sterilizer is opened, when quality of water does not reach preset standard for lasting long time reach second preset duration, send water storage box self-cleaning equipment to open the instruction to this pump house water storage box self-cleaning equipment in the water storage box, so that water storage box self-cleaning equipment opens, first preset duration is less than second preset for long time.
Preferably, the real-time water quality data is real-time water quality, and the step of determining whether the water quality in the water storage tank of the pump room reaches a preset standard according to the real-time water quality data in the real-time data includes:
and judging whether the real-time water quality is less than the preset water quality, if so, determining that the water quality in the water storage tank of the pump room reaches the preset standard, and if not, determining that the water quality of the water stored in the pump room does not reach the preset standard.
Compared with the prior art, the invention has the following beneficial effects:
through the whole optimization research and development design of pump room, water storage tank, water pump unit, water supply pipe assembly, data acquisition and transmission unit, pump room monitoring unit, wisdom water tank regulation and control unit, water conservancy data monitoring unit, water quality monitoring and processing unit, pipe network water conservancy data monitoring unit, cloud platform server and the big data analysis module of cloud platform, an energy-conserving peak-shifting wisdom water supply system based on cloud has been manufactured. The pump room monitoring unit of each pump room in the water supply system can send real-time data to the cloud platform big data analysis module through the data acquisition and transmission unit of the pump room, each intelligent water tank regulation and control unit sends the maximum water storage amount of the corresponding water storage tank to the cloud platform big data analysis module, and the cloud platform big data analysis module determines the target water storage amount of the water storage tank corresponding to the real-time data based on a pre-trained target neural network model; then based on real-time water storage capacity, the biggest water demand of the corresponding water storage box that wisdom water tank regulation and control unit confirmed, whether target water storage capacity and wisdom water tank regulation and control unit have begun to supply water for the water storage box, supply water to the water storage box when being used for controlling this wisdom water tank regulation and control unit before real-time water storage capacity reaches target water storage capacity, realize when the non-water peak, optimize distribution pipe network water supply resource and carry out the overall regulation and storage to the water storage box, reduce the instant water demand that the water supply network supplied water when the water peak, thereby realize the purpose that the off-peak supplied water. And further, the optimized utilization of water supply network facilities is realized, and the saving of national water supply resources and supporting facilities is realized. However, it is not necessary for any product or method of practicing the invention to achieve all of the above-described advantages simultaneously. The method has the following specific beneficial effects:
1) the pump room monitoring unit of each pump room sends real-time data to the cloud platform big data analysis module through the data acquisition and transmission unit of the pump room, and the cloud platform big data analysis module determines the target water storage capacity of the water storage tank corresponding to the real-time data based on a pre-trained target neural network model; then based on the real-time water storage capacity, the maximum water demand of the corresponding water storage tank determined by the intelligent water tank regulation and control unit, the target water storage capacity and whether the intelligent water tank regulation and control unit starts to supply water to the water storage tank or not, the intelligent water tank regulation and control unit is used for controlling the water storage tank to supply water before the real-time water storage capacity reaches the target water storage capacity so as to meet the requirement of optimizing and distributing a pipe network water supply resource to carry out overall regulation and storage on the water tank at the non-water peak time, reduce the instant water demand of water supply of the municipal water supply network at the water peak time, shorten the pressure fluctuation range of the municipal water supply network and the water supply pipeline assembly, reduce the pressure fluctuation of the municipal water supply network and the water supply pipeline assembly, avoid the pressure fluctuation of the municipal water supply network from increasing the instant water supply pressure of the water plant, reduce the leakage loss of the pipe network, ensure the, the saving of national water supply resources and supporting facilities is realized;
2) the intelligent water tank regulation and control unit can be connected with the cloud platform server, receives an instruction of overall water storage of the cloud platform server and carries out peak shifting water storage on the pump house water storage tank;
3) the redundancy of the real-time data is removed based on the discrete cosine transform matrix, so that the accuracy of the subsequent predicted target water storage amount is improved, and the convenience of calculation of the subsequent predicted target water storage amount can be improved by carrying out normalization processing on the redundancy-removed real-time data;
4) the initial target neural network model is obtained by training the initial target neural network model, and then the initial neural network model is trained in a matched training mode, so that the target neural network model which enables real-time water consumption sample data of a user to be associated with the sample water storage amount of the corresponding water storage tank can be obtained, the target water storage amount can be predicted through the target neural network model, so that water pumping or water pumping stopping can be conveniently carried out on the corresponding water pump unit based on the target water storage amount and the corresponding water storage tank, the purpose of reducing energy waste is achieved, the water pump unit does not need to be controlled manually, the purpose of automatically controlling the energy-saving storage regulation operation of the water supply system is achieved, the labor cost is saved, and the working efficiency is improved;
5) the equation set relation established between the input layer and the output layer is an excitation function, the excitation function can exclude user real-time water consumption sample data with more than negative excitation coefficient from user real-time water consumption sample data, generally, in the actual use process of the water supply system, the user real-time water consumption sample data with more than negative excitation coefficient in the user real-time water consumption sample data is considered to be data with excessive noise, the user real-time water consumption sample data with excessive noise is excluded, the adverse effect of noise on the stability of a target neural network model in the water supply system can be reduced or eliminated, and meanwhile, the prediction error of the target neural network model in the water supply system can be reduced.
6) The water storage tank of each pump room is internally provided with a sterilizer and a water storage tank self-cleaning device, and in the mode, the sterilizer and the water storage tank self-cleaning device are controlled or instructed to sterilize and clean the water body in the water storage tank when the water quality of the corresponding water storage tank in the pump room does not reach the preset standard so as to ensure that the water supply quality of a user is safe and reaches the standard.
Drawings
FIG. 1 is a schematic diagram of a simulation structure of connection among a cloud platform server of a cloud platform system, a cloud platform big data analysis module embedded in the cloud platform server, and a water storage tank, a data acquisition and transmission unit, a pump room monitoring unit, a smart water tank regulation and control unit, a hydraulic data monitoring unit, a water quality monitoring and processing unit, and a pipe network hydraulic data monitoring unit which are arranged in each pump room in the water supply system according to the present invention;
fig. 2 is a schematic view of a virtual network connection structure between the whole area of each pump room and a cloud platform system provided in the water supply system according to the present invention;
FIG. 3 is a schematic structural diagram of the interconnection between the specific hardware set of a pump room in the water supply system and the components in the cloud platform system;
FIG. 4 is a schematic diagram of a connection structure of an internal training process of a cloud platform system target neural network model in the water supply system.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the drawings and the accompanying description. It is to be understood that the described embodiments are merely a few embodiments of the invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive effort based on the embodiments of the present invention, are within the scope of the present invention.
It should be noted that the terms "comprises" and "comprising" and any variations thereof in the embodiments of the present invention and the description of the drawings are intended to cover non-exclusive inclusions. A process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but preferably also includes other steps or elements not listed or inherent to such process, method, article, or apparatus.
The invention discloses an energy-saving peak-shifting intelligent water supply system based on cloud technology, which can comprehensively control the water storage amount in a water storage tank and a water storage tank in each pump room in a control area through a cloud platform consisting of a cloud platform server and a matched network infrastructure, realize the optimization of a distribution pipe network water supply resource to carry out the overall regulation and storage on the water tank at the non-water peak time, reduce the instant water demand of the water supply of a municipal water supply pipe network at the water peak time, shorten the pressure fluctuation range of the municipal water supply pipe network and a water supply pipe assembly, and has small pressure fluctuation of the municipal water supply pipe network and the water supply pipe assembly, avoid municipal water supply pipe network's pressure fluctuation to increase the instantaneous water supply pressure of water plant simultaneously, reduce the leakage of pipe network, guaranteed water supply system even running of water plant, reduced the demand of rebuilding of water plant municipal water supply pipe network and water supply pipe assembly, realized saving of national water supply resource and supporting facility. The following detailed description of embodiments of the invention is described in conjunction with the accompanying drawings, in which:
FIG. 1 is a schematic diagram of a simulation structure of connection among a cloud platform server of a cloud platform system, a cloud platform big data analysis module embedded in the cloud platform server, and a water storage tank, a data acquisition and transmission unit, a pump room monitoring unit, a smart water tank regulation and control unit, a hydraulic data monitoring unit, a water quality monitoring and processing unit, and a pipe network hydraulic data monitoring unit which are arranged in each pump room in the water supply system according to the present invention; fig. 1 specifically includes: the water supply system comprises a cloud platform server (1) of the cloud platform system, a cloud platform big data analysis module (2) embedded in the cloud platform server (1), each pump room (3) controlled by the cloud platform server (1), a water storage tank (4) arranged inside each pump room (3), a data acquisition and transmission unit (5), a pump room monitoring unit (6), a smart water tank regulation and control unit (7), a hydraulic data monitoring unit (8), a water quality monitoring and processing unit (9) and a pipe network hydraulic data monitoring unit (10).
In order to more vividly show and illustrate the virtual network connection between the whole area of each pump room and the cloud platform system arranged in the water supply system of the invention, a figure 2 representation (in particular, see figure 2) is used; to more visually illustrate the interconnections between the specific hardware components of a pump house and the components of a self-contained cloud platform system in a water supply system of the present invention, a representation of fig. 3 (see fig. 3 in particular) is used; the structure of the energy-saving peak-shifting intelligent water supply system based on the cloud technology and the connection relationship between specific supporting equipment in each pump room (3) controlled by the cloud platform server (1) can be more visually represented and explained by fig. 2 and 3.
In the actual use and operation process, in order to intelligently control the intelligent water tank regulation and control unit (7) in each pump room (3) arranged in each area of the water supply system in a city, the cloud platform server (1) for the pump rooms (3) in the area is arranged in a matched mode and used for controlling specific matched equipment in each pump room (3) to achieve energy-saving peak-shifting intelligent water supply operation, and the cloud platform server (1) can be arranged in a master control center of the whole city water supply system. This cloud platform server (1) is embedded to have cloud platform big data analysis module (2), and cloud platform big data analysis module (2) can carry out communication connection through wireless local area network and the concrete corollary equipment of setting in the pump house. The intelligent water tank regulation and control unit (7) of each pump room (3) sends the maximum water storage amount of the water storage tank (4) corresponding to the regulation and control unit to the cloud platform big data analysis module (2);
the water storage tank (4) that sets up in every pump house (3) to and the wisdom water tank regulation and control unit (7) that corresponds with this water storage tank (4), municipal water supply pipe network and water supply pipe assembly can be controlled in this wisdom water tank regulation and control unit (7), impounds or does not impound to water storage tank (4), and water storage tank (4) can save the water that supplies water to the user, and water storage tank (4) can supply water to different users through a plurality of water supply pipe assemblies. Meanwhile, a security door is arranged on the water storage tank (4), and the security door is used for realizing the security protection of the water storage tank (4);
the intelligent water tank regulating and controlling unit (7) of each pump room (3) can also acquire the real-time water storage capacity of the corresponding water storage tank (4) in the pump room (3) and send the acquired real-time water storage capacity to the pump room monitoring unit (6) of the pump room (3); meanwhile, the water storage tank (4) is started or stopped to store water according to an instruction from the cloud platform big data analysis module (2) which is sent to the intelligent water tank regulation and control unit (7) through the pump room monitoring unit (6);
the hydraulic data monitoring unit (8) of each pump room (3) can acquire hydraulic data of water supplied to a user by the pump room (3) and comprises water supply pressure and water supply flow, the water supply flow is the water consumption of the real-time user, and the data are sent to the pump room monitoring unit (6);
the water quality monitoring and processing unit (9) of each pump room (3) can acquire real-time water quality data of the corresponding water storage tank (4) in the pump room (3) and send the acquired real-time water quality data to the pump room monitoring unit (6) of the pump room (3);
the pipe network hydraulic data monitoring unit (10) of each pump room (3) can acquire hydraulic data of a municipal water supply pipe network entering the tail end of the pump room (3) and comprises water supply pressure and water supply flow, the water supply flow is the real-time water supply amount of the municipal water supply pipe network to the corresponding water storage tank (4) in the pump room (3), and the data are sent to the pump room monitoring unit (6);
the pump room monitoring unit (6) of each pump room (3) can receive real-time water supply pressure data, water supply flow data (namely real-time user water consumption data), real-time water quality data and hydraulic data, entering the tail end of the pump room, of the pump room (3) sent by the intelligent water tank regulating and controlling unit (7), the hydraulic data monitoring unit (8), the water quality monitoring and processing unit (9) and the pipe network hydraulic data monitoring unit (10) of the pump room (3) and supplying water to users, wherein the real-time water supply pressure data, the water supply flow data (namely real-time water supply amount of the water storage tank (4) corresponding to the pump room (3) from the municipal water supply network) comprise water supply pressure and water supply flow data, and the real-time data are transmitted to the cloud platform big data analysis module (2) through the data collecting and transmitting unit (5);
the cloud platform big data analysis module (2) can receive real-time water supply pressure data, water supply flow data (namely real-time user water consumption data), real-time water quality data and hydraulic data of a municipal water supply network entering the tail end of the pump room, wherein the real-time water supply pressure data, the real-time water quality data and the water supply flow data are sent by the intelligent water tank regulation and control unit (7), the hydraulic data monitoring unit (8), the water quality monitoring and processing unit (9) and the pipe network hydraulic data monitoring unit (10) in each pump room (3), and the real-time water supply pressure data and the real-time water supply flow data (namely the real-time water supply amount of the municipal water supply network to the corresponding water storage tank (4) in the pump room;
because the water consumption of the real-time user of each pump room (3) has certain periodicity and regularity, namely the water storage amount of the corresponding water storage tank (4) in each pump room (3) only needs to meet the water demand of the user in the next time period; the water storage amount of the water storage tank (4) in full water storage is not required to be reached, when the water storage amount of the water storage tank (4) meets the water demand of a user, the cloud platform big data analysis module (2) can control the corresponding intelligent water tank regulation and control unit (7) in the region to stop storing water for the corresponding water storage tank (4). Therefore, after the cloud platform big data analysis module (2) receives the real-time data of each pump room (3), the target water storage amount of the water storage tank (4) corresponding to the real-time data needs to be predicted according to the real-time data of each pump room (3).
In order to predict the target water storage amount of the water storage tank (4) corresponding to the real-time data, a target neural network model is trained in the water supply system in advance. The target neural network model in the water supply system of the invention is as follows: based on the real-time water consumption of the user, the real-time water consumption sample data of the user and the sample water storage amount of the corresponding water storage tank (4) as training data of the model, an initial target neural network model is trained to obtain the initial neural network model, the target neural network model can enable the real-time water consumption sample data of the user to be associated with the sample water storage amount of the corresponding water storage tank (4), and the real-time water consumption sample data of the user comprises the real-time sample water storage amount of the corresponding water storage tank (4), the real-time sample user water consumption, the accumulated sample water supply amount, the real-time sample water quality data and the real-time sample water supply.
The training process of the target neural network model is specifically introduced as follows:
the training process of the target neural network model can be:
acquiring sample data of real-time water consumption of a user for training, wherein the sample data comprises the real-time sample water storage amount of a corresponding water storage tank (4), the real-time sample user water consumption, the accumulated sample water supply amount, the real-time sample water quality data and the real-time sample water supply amount;
inputting sample data and real-time user water consumption into an initial neural network model, wherein the initial neural network model comprises an input layer, a hidden layer, an output layer and an equation set relation established between the input layer and the output layer, and the equation set relation established between the input layer and the output layer is an excitation function;
according to the sample data and the sample water storage capacity of the corresponding water storage tank (4), the weight coefficient between each node of the input layer and each node of the hidden layer, the weight coefficient between each node of the hidden layer and each node of the output layer and an excitation function for establishing an equation set relation between the input layer and the output layer;
solving the equation set to obtain the values of the weight coefficients between each node of the input layer and each node of the hidden layer and the values of the weight coefficients between each node of the hidden layer and the nodes of the output layer;
substituting the values of the weight coefficients between each node of the input layer and each node of the hidden layer, the values of the weight coefficients between each node of the hidden layer and each node of the output layer and the sample data of the real-time water consumption of the user into an equation set to solve the initial water storage capacity of the water storage tank (4) corresponding to the real-time water consumption of the user;
calculating a difference value between the initial water storage amount and the target water storage amount;
when the difference value is larger than the preset difference threshold value, adjusting the first weight and the second weight based on the difference value, and returning to execute the step of inputting the real-time water consumption sample data of the user and the sample water storage amount of the corresponding water storage tank (4) into the initial neural network model; the first weight coefficient is a first weight coefficient between the first input layer node and the first hidden layer node, and the second weight coefficient is a second weight coefficient between the second input layer node and the first hidden layer node;
and when the difference value is not greater than the preset difference value threshold value, finishing training to obtain a target neural network model containing the corresponding relation between the real-time water consumption sample data of the user and the sample water storage amount of the corresponding water storage tank (4).
When a target neural network model is established, sample data of user real-time water consumption used in a training set and sample water storage amount of a corresponding water storage tank (4) need to be acquired, wherein the sample data of the user real-time water consumption comprises the sample real-time water storage amount of the corresponding water storage tank (4), the sample real-time user water consumption, sample accumulated water supply amount, sample real-time water quality data and sample real-time water supply amount; the sample data of the real-time water consumption of the users used in the training set and the sample water storage amount of the corresponding water storage tank (4) are the data collected in different time aiming at different pump rooms (3).
The invention relates to an initial neural network model in an energy-saving peak-shifting intelligent water supply system based on a cloud technology, which can be understood as follows: the related electronic equipment of the specific corollary equipment in the invention needs to construct an initial neural network model, and train the initial neural network model to obtain a target neural network model. In one of the ways that can be realized, an initial neural network model comprising an input layer, a hidden layer, an output layer and an equation system excitation function relation established between the input layer and the output layer is constructed by using a coffee tool;
after the system acquires the real-time water consumption sample data of the user and the corresponding sample water storage capacity, the real-time water consumption data of the user and the real-time water storage capacity of the corresponding water storage tank (4) can be input into an initial neural network model, and the initial neural network model comprises an input layer, a hidden layer, an output layer and an equation set excitation function relation established between the input layer and the output layer;
then, establishing an equation set according to the water consumption of a real-time user, the real-time water storage amount of the corresponding water storage tank (4), the weight coefficient between each node of the input layer and each node of the hidden layer, the weight coefficient between each node of the hidden layer and each node of the output layer and the excitation function relationship of the equation set between the input layer and the output layer, wherein at the moment, the weight coefficient between each node of the input layer and each node of the hidden layer and the weight coefficient between each node of the hidden layer and each node of the output layer are unknown numbers;
after the equation set is established, solving the equation set to obtain the values of the weight coefficients between each node of the input layer and each node of the hidden layer and the values of the weight coefficients between each node of the hidden layer and the nodes of the output layer; substituting the values of the weight coefficients between each node of the input layer and each node of the hidden layer, the values of the weight coefficients between each node of the hidden layer and each node of the output layer and the sample data of the real-time water consumption of the user into an equation set to solve the initial water storage capacity of the water storage tank (4) corresponding to the real-time water consumption of the user;
calculating a difference value between the initial water storage amount of the corresponding water storage tank (4) and the sample water storage amount through a predefined objective function, and when the difference value is larger than a preset difference threshold value, indicating that the initial neural network model at the moment cannot be suitable for most of real-time water consumption sample data of users, adjusting a first weight coefficient and a second weight coefficient through a back propagation method according to the difference value, and returning to execute the step of inputting the real-time water consumption sample data of the users and the sample water storage amount of the corresponding water storage tank (4) into the initial neural network model;
in the training process, the system can circularly traverse all the real-time water consumption sample data of the users in each area of the water supply system in the city, the first weight coefficient and the second weight coefficient are continuously adjusted, when the difference value is not greater than the preset difference threshold value, the initial neural network model at the moment can be suitable for the real-time water consumption sample data of most users, an accurate result is obtained, at the moment, the training is completed, and the target neural network model containing the corresponding relation between the real-time water consumption sample data of the users and the sample water storage amount of the corresponding water storage tank (4) is obtained.
The invention relates to an initial neural network model in an energy-saving peak-shifting intelligent water supply system based on a cloud technology, which can be concretely understood as follows: the input layer comprises a first input layer node, a second input layer node and a third input layer node, the hidden layer comprises a first hidden layer node, a second hidden layer node and a third hidden layer node, the output layer comprises an output layer node, and the steps of according to the real-time water consumption sample data of a user and the sample water storage amount of a corresponding water storage tank (4), the weight coefficient between each node of the input layer and each node of the hidden layer, the weight coefficient between each node of the hidden layer and the output layer node of the output layer and the set of equation excitation functions between the input layer and the output layer can comprise:
establishing a first linear equation according to the real-time water consumption sample data of the user, a first weight coefficient between a first input layer node and a first hidden layer node, a second weight coefficient between a second input layer node and the first hidden layer node and a first constant between a third input layer node and the first hidden layer node;
establishing a second linear equation according to the real-time water consumption sample data of the user, a third weight coefficient between the first input layer node and the second hidden layer node, a fourth weight coefficient between the second input layer node and the second hidden layer node and a second constant between the third input layer node and the second hidden layer node;
establishing a third linear equation according to the real-time water consumption sample data of the user, a fifth weight coefficient between the first input layer node and the third hidden layer node, a sixth weight coefficient between the second input layer node and the third hidden layer node and a third constant between the third input layer node and the third hidden layer node;
establishing a fourth function equation according to an equation set excitation function relation established between the input layer and the output layer, the sample water storage amount corresponding to the real-time water consumption sample data of the user, the function value of the first linear equation, the function value of the second linear equation and the function value of the third linear equation, a seventh weight coefficient between the first hidden layer node and the output layer node, an eighth weight coefficient between the second hidden layer node and the output layer node and a ninth weight coefficient between the third hidden layer node and the output layer node;
simultaneously establishing a first linear equation, a second linear equation, a third linear equation and a fourth function equation to obtain an equation set between the input layer and the output layer;
solving the equation set to obtain the values of the weight coefficients between the nodes of the input layer and the nodes of the hidden layer and the values of the weight coefficients between the nodes of the hidden layer and the nodes of the output layer, wherein the method comprises the following steps:
solving the equation set to obtain a value of a first weight, a value of a second weight, a value of a third weight, a value of a fourth weight, a value of a fifth weight, a value of a sixth weight, a value of a seventh weight, a value of an eighth weight and a value of a ninth weight;
substituting the value of the weight coefficient between each node of the input layer and each node of the hidden layer, the value of the weight coefficient between each node of the hidden layer and each node of the output layer and the user real-time water consumption sample data into an equation set, and solving the initial water storage capacity of the water storage tank (4) corresponding to the user real-time water consumption sample data, wherein the method comprises the following steps of:
and substituting the value of the first weight, the value of the second weight, the value of the third weight, the value of the fourth weight, the value of the fifth weight, the value of the sixth weight, the value of the seventh weight, the value of the eighth weight, the value of the ninth weight and the user real-time water consumption sample data into the equation set to solve the initial water storage capacity of the water storage tank (4) corresponding to the user real-time water consumption sample data.
In an exemplary embodiment (see fig. 4), fig. 4 is a schematic diagram of a connection structure of an internal training process of a cloud platform system target neural network model in a water supply system according to the present invention. As can be seen from fig. 4, three nodes X1 on the left side of the target neural network model are first input layer nodes, X2 are second input layer nodes and 1 is a third input layer node, a node y on the right side is an output layer node, h1 is a first hidden layer node, h2 is a second hidden layer node, h3 is a third hidden layer node, and σ represents an excitation function, each node is also called a neuron, the neurons are connected to form the target neural network, corresponding weight coefficients and constants exist between the input layer and the hidden layer, corresponding weight coefficients exist between the hidden layer and the output layer, the target neural network model is determined by an excitation function (excitation function relationship of an equation set established between the input layer and the output layer), the weight coefficients, constants and connection modes between the neurons, the connection modes of the target neural network model are fully connected, that is, that any two nodes between two adjacent layers are connected, this has the advantage of being able to mine potential correlations between different types of data.
The relationship between the input and output of the target neural network model in fig. 4 can be defined by the following formula:
a1＝w1-11x1+w1-21x2+b1-1
a2＝w1-12x1+w1-22x2+b1-2
a3＝w1-13x1+w1-23x2+b1-3
y＝σ(w2-1σ(a1)+w2-2σ(a2)+w2-3σ(a3))
Figure GDA0002380155090000231
wherein, a1Is a function value of a first linear equation, a2Is the function value of a second linear equation, a3As a function of a third linear equation, w1-11Is a first weight coefficient, w, between a first input layer node and a first hidden layer node1-21Is a second weight coefficient, w, between the second input layer node and the first hidden layer node1-12Is a third weight coefficient, w, between the first input layer node and the second hidden layer node1-22Is a fourth weight coefficient, w, between the second input layer node and the second hidden layer node1-13Is a fifth weight coefficient, w, between the first input layer node and the third hidden layer node1-23Is a sixth weight coefficient, w, between the second input layer node and the third hidden layer node2-1Is a seventh weight coefficient, w, between the first hidden layer node and the output layer node2-2Is an eighth weight coefficient, w, between the second hidden layer node and the output layer node2-3Is a ninth weight coefficient between the third hidden layer node and the output layer node, b1-1As a first constant between the third input layer node and the first hidden layer node, b1-2As a second constant between the third input layer node and the second hidden layer node, b1-3Is a third constant between the third input layer node and the third hidden layer node, y is the sample water storage capacity corresponding to the real-time water consumption sample data of the user, and sigma (x)i) To relate to xiExcitation function, xiReal-time water sample data, x, for the ith user1Real-time water sample data, x, for the 1 st user2Real-time water sample data for the 2 nd user, a1＝w1-11x1+w1-21x2+b1-1Is a first linear equation of2＝w1- 12x1+w1-22x2+b1-2Is a second linear equation a3＝w1-13x1+w1-23x2+b1-3Is a third linear equation, y ═ σ (w)2-1σ(a1)+w2-2σ(a2)+w2-3σ(a3) Is a fourth function equation.
The function values of the excitation function in the above exemplary example can be: the sum of the excitation coefficient and a target maximum value, wherein the target maximum value is the maximum value in the real-time water consumption sample data of the user and a target product, and the target product is the product of the excitation coefficient and the real-time water consumption sample data of the user. The excitation coefficient in the above formula is-0.1, and the real-time water consumption sample data of the user is the real-time water consumption sample data of the ith user. The excitation function in the embodiment of the invention can eliminate the user real-time water consumption sample data which is larger than the negative excitation coefficient in the user real-time water consumption sample data, generally, in the actual use process of the water supply system, the user real-time water consumption sample data which is larger than the negative excitation coefficient in the user real-time water consumption sample data is considered to be data with overlarge noise, and the user real-time water consumption sample data with overlarge noise is eliminated, so that the adverse effect of noise on the stability of a target neural network model in the water supply system can be reduced or eliminated, and meanwhile, the prediction error of the target neural network model in the water supply system can be reduced.
The following table shows the comparison of the performance of the excitation function in the embodiment of the present invention and the standard ReLU (Rectified Linear Unit) excitation function in predicting the target water storage capacity, where the smaller the value in the table represents the average error value, the higher the performance, and it can be seen that the prediction error is smaller and the accuracy is higher when the excitation function in the embodiment of the present invention is used for predicting the target water storage capacity.
Excitation function Data set 1 Data set 2 Data set 3
Examples of the invention 5.2％ 2.3％ 10.1％
Standard ReLU 10.5％ 5.5％ 13.7％
In conclusion, the initial neural network model is trained through the training mode, a target neural network model which enables real-time water consumption sample data of a user to be associated with the sample water storage amount of the corresponding water storage tank can be obtained, the target water storage amount of the corresponding water storage tank can be predicted through the target neural network model, and the water consumption regularity characteristic of the user can be obtained, so that the water pumping of the water pump unit in the pump room is controlled to be carried out or stopped based on the target water storage amount of the corresponding water storage tank, the purpose of reducing energy waste is achieved, meanwhile, the water pump unit does not need to be controlled manually, the purpose of remote automatic control (namely cloud technology control) is achieved, labor cost is saved, and working efficiency is improved.
According to the invention, a cloud platform big data analysis module (2) in a water supply system determines the target water storage capacity of a water storage tank (4) corresponding to real-time data based on a pre-trained target neural network model for the real-time data of each pump room (3), when the real-time water storage capacity of the corresponding water storage tank (4) in the real-time data does not reach the target water storage capacity and a smart water tank regulation and control unit (7) of the pump room (3) is not opened to store water for the water storage tank (4), the fact that the water storage capacity in the water storage tank (4) is not enough for a user corresponding to the pump room (3) is indicated, and the smart water tank regulation and control unit (7) is required to indicate and control a municipal network and a water supply pipeline component to store water for the water storage tank (4; at the moment, the cloud platform big data analysis module (2) sends a starting instruction to the intelligent water tank regulation and control unit (7) of the pump room (3), so that the intelligent water tank regulation and control unit (7) of the pump room (3) instructs the municipal water supply network and the water supply pipeline assembly to start to store water to the water storage tank (4);
in order to save energy, when the water storage amount in the water storage tank (4) reaches the target water storage amount, the water storage to the water storage tank (4) is stopped. At this moment, the cloud platform big data analysis module (2) can calculate the first target water storage time based on the real-time water storage amount of the corresponding water storage tank (4) in the real-time data, the maximum water demand and the target water storage amount of the water storage tank (4) stored by the intelligent water tank regulation and control unit (7) of the pump room (3), when the first target water storage time is reached, the cloud platform big data analysis module (2) sends a first stop instruction to the intelligent water tank regulation and control unit (7) of the pump room (3), so that the intelligent water tank regulation and control unit (7) of the pump room (3) stops storing water to the corresponding water storage tank (4).
Wherein, based on the real-time water storage capacity of the corresponding water storage tank (4) in the real-time data, the maximum water storage capacity and the target water storage capacity of the water storage tank (4) stored by the intelligent water tank regulation and control unit (7) of the pump room (3), the first target water storage time can be calculated as follows: and calculating the difference value between the target water storage amount and the real-time water storage amount, calculating the quotient of the difference value and the rated water pumping amount, and taking the quotient as the first target water storage time.
When the real-time water storage amount of the corresponding water storage tank (4) in the real-time data does not reach the target water storage amount and the intelligent water tank regulation and control unit (7) of the pump room (3) is opened to store water to the corresponding water storage tank (4), the fact that the water storage amount in the water storage tank (4) is not enough for a user in the area where the pump room (3) is located is described, and the intelligent water tank regulation and control unit (7) is required to open the corresponding municipal water supply network and water supply pipeline assembly to store water to the corresponding water storage tank (4); because the intelligent water tank regulation and control unit (7) of the pump room (3) is opened to store water in the corresponding water storage tank (4), in order to improve the update rate of the water body in the corresponding water storage tank (4), the water storage in the corresponding water storage tank (4) can be stopped when the water storage amount in the corresponding water storage tank (4) reaches the target water storage amount. At the moment, the cloud platform big data analysis module (2) can calculate second target water storage time based on the real-time water storage amount in the real-time data, the maximum water demand of the corresponding water storage tank (4) stored by the intelligent water tank regulation and control unit (7) of the pump room (3) and the target water storage amount, when the second target water storage time is reached, the cloud platform big data analysis module (2) sends a second stop instruction to the intelligent water tank regulation and control unit (7) of the pump room (3) so that the intelligent water tank regulation and control unit (7) in the pump room (3) stops the corresponding municipal water supply network and water supply pipeline assembly and stores water to the corresponding water storage tank (4), wherein the mode of calculating the second target water storage time is the same as the mode of calculating the first target water storage time, and detailed description is omitted here.
When the real-time water storage amount of the corresponding water storage tank (4) in the real-time data reaches the target water storage amount and the intelligent water tank regulation and control unit (7) is opened to store water into the corresponding water storage tank (4), the fact that the water storage amount in the corresponding water storage tank (4) is enough for a user of the pump room (3) is indicated, and the intelligent water tank regulation and control unit (7) is not required to continuously keep the opened state to store water into the water storage tank (4) any more; at this moment, cloud platform big data analysis module (2) can send the third and stop instruction to wisdom water tank regulation and control unit (7) of this pump house (3) to thereby make wisdom water tank regulation and control unit (7) of this pump house (3) close and stop municipal water supply network and the water supply pipe assembly that corresponds, impound to corresponding water storage box (4).
Through setting up wisdom water tank regulation and control unit (7), water conservancy data monitoring unit (8), water quality monitoring and processing unit (9), pipe network water conservancy data monitoring unit (10) in every pump house (3) can acquire the accurate water storage capacity, the water quality data of corresponding water storage box (4) and user's real-time water consumption in real time. Meanwhile, target water storage amount required by the water storage tanks (4) corresponding to all the areas is predicted based on the target neural network model, and the water use regularity characteristics of the user are obtained; according to the water consumption regularity characteristics of the users, the water storage tanks (4) are regulated and stored, so that specific matched water supply system facilities can store water to the corresponding water storage tanks (4), the water is stored to the required target water storage amount of each pump room (3) user before the water consumption peak comes, and the pressure of a water supply network is prevented from being greatly fluctuated due to the increase of the instant water consumption of the users when the water consumption peak comes; at the moment, the regulation and storage capacities of all the water storage tanks (4) in the management area can be comprehensively regulated through the cloud platform system technology, so that the peak load shifting regulation and storage of the municipal water supply network can be realized; municipal water supply network and water supply pipe assembly pressure fluctuation scope can be shortened through the regulation water supply of peak staggering, municipal water supply network and water supply pipe assembly pressure fluctuation are little, avoid municipal water supply network's pressure fluctuation to increase the instantaneous water supply pressure of water factory simultaneously, reduce the leakage loss of municipal water supply network and water supply pipe assembly, water supply system even running of water factory has been guaranteed, the demand of rebuilding of water factory municipal water supply network and water supply pipe assembly has been reduced, the saving of national water supply resource and supporting facility has been realized and has been practiced thrift.
In addition, the real-time data in the water supply system is huge, and most of the real-time data are redundant data without reference value. In order to improve the accuracy of the subsequent target water storage amount prediction, the cloud platform big data analysis module (2) can be used for performing discrete cosine transform on the real-time data based on a discrete cosine transform matrix to obtain the redundancy-removed real-time data before determining the target water storage amount of the water storage tank (4) corresponding to the real-time data based on a pre-trained target neural network model.
The discrete cosine transform of the real-time data based on the discrete cosine transform matrix to obtain the redundancy-removed real-time data comprises the following steps:
arranging the real-time water storage amount, the real-time user water consumption amount, the accumulated water supply amount, the real-time water quality data and the sampling point data contained in the real-time water supply amount of the corresponding water storage tank (4) in the real-time data into a first matrix, calculating the product of the first matrix and the discrete cosine transform matrix to obtain a second matrix, and taking each element in the second matrix as the real-time data after redundancy removal.
An exemplary embodiment of the present invention, in which the water supply system performs discrete cosine transform on the real-time data based on the discrete cosine transform matrix, is as follows: the real-time data includes multi-channel data (replaced by M-channel data), and in the invention, the real-time data includes real-time water storage amount of the water storage tank, real-time user water consumption amount, accumulated water supply amount, real-time water quality data and real-time water supply amount. At this time, M may be 5, and assuming that each path of data includes N sampling points, the real-time data is arranged into a first matrix a with N rows and M columns, where a path of data includes samplesThe dot data constitute a column vector c of the first matrix AjFor the column vector c of the first matrix AjDefining the discrete cosine transform matrix U as an N × N orthogonal matrix such that:
dj＝Ucj,j＝1，2,…,N,
Figure GDA0002380155090000281
in the above formula: djIs the column vector of the second matrix, U is the discrete cosine transform matrix, cjIs a column vector of the first matrix, j is the number of rows of the second matrix, N is the number of columns of the second matrix, upqIs the element of the p-th row and the q-th column in the discrete cosine transform matrix U, p is the row number of the discrete cosine transform matrix, and q is the column number of the discrete cosine transform matrix.
After the second matrix is obtained, each element in the second matrix is used as the real-time data after redundancy removal, so that the redundancy removal is carried out on the real-time data based on the discrete cosine transform matrix, and the accuracy of the subsequent target water storage capacity prediction is improved.
Because the calculated amount of the normalized data is small and the normalized data is convenient to calculate relative to the non-normalized data, after the real-time data is subjected to redundancy removal, the calculation is convenient for the subsequent prediction of the target water storage amount, the normalized real-time data can be subjected to normalization processing, namely, the cloud platform big data analysis module (2) calculates the mean value and the standard deviation of the real-time data subjected to redundancy removal, the normalized real-time data is obtained by performing normalization processing on the real-time data subjected to redundancy removal based on the mean value and the standard deviation, and the target water storage amount of the water storage tank (4) corresponding to the normalized real-time data is determined based on the target neural network model which is trained in advance.
The method comprises the following steps that a specific cloud platform big data analysis module (2) calculates the mean value and the standard deviation of real-time data after redundancy removal, and normalization processing is carried out on the real-time data after redundancy removal based on the mean value and the standard deviation to obtain normalized real-time data, and the method comprises the following steps:
the cloud platform big data analysis module (2) is specifically used for calculating a first mean value and a first standard deviation of first elements representing real-time water storage capacity of the corresponding water storage tanks (4) in the second matrix, calculating a first difference value between each first element and the first mean value, and calculating a first quotient value between the first difference value and the first standard deviation;
calculating a second mean value and a second standard deviation of second elements representing real-time user water consumption in the second matrix, calculating a second difference value of each second element and the second mean value, and calculating a second quotient value of the second difference value and the second standard deviation;
calculating a third mean value and a third standard deviation of third elements representing the cumulative water supply in the second matrix, calculating a third difference value of the third elements and the third mean value for each third element, and calculating a third quotient of the third difference value and the third standard deviation;
calculating a fourth mean value and a fourth standard deviation of a fourth element representing real-time water quality data in the second matrix, calculating a fourth difference value of the fourth element and the fourth mean value for each fourth element, and calculating a fourth quotient value of the fourth difference value and the fourth standard deviation;
calculating a fifth mean value and a fifth standard deviation of fifth elements representing the real-time water supply amount in the second matrix, calculating a fifth difference value of each fifth element and the fifth mean value, and calculating a fifth quotient value of the fifth difference value and the fifth standard deviation;
and taking the first quotient value, the second quotient value, the third quotient value, the fourth quotient value and the fifth quotient value as normalized real-time data.
And normalizing the first element representing the real-time water storage amount in the second matrix by the following formula:
Figure GDA0002380155090000301
wherein, X*Is a first quotient, X is a first element, μ is a first mean, and δ is a first standard deviation.
Similarly, the formula for normalizing the second element, the third element, the fourth element and the fifth element in the second matrix is the same as the first element, and will not be described in detail here.
Therefore, the calculation convenience of the subsequent predicted target water storage amount is improved in a mode of carrying out normalization processing on the redundancy-removed real-time data.
When the water quality in the water storage tank (4) is poor and not up to the standard, the cloud platform big data analysis module (2) does not control specific matched water supply system facilities to supply water to users. At the moment, a sterilizer and water tank self-cleaning equipment are installed in a water storage tank (4) corresponding to water supply of a user in each pump room (3), a cloud platform big data analysis module (2) can determine whether the water quality in the water storage tank (4) corresponding to the water supply of the user in the pump room (3) reaches a preset standard according to real-time water quality data in the real-time data, when the duration time that the water quality in the water storage tank (4) does not reach the preset standard reaches a first preset time, the cloud platform big data analysis module (2) sends a sterilizer starting instruction to a sterilizer in the water storage tank (4) corresponding to the water supply of the user in the pump room (3), and the sterilizer is started to sterilize the water in the water storage tank (4); when the duration that the water quality in the water storage tank (4) does not reach the preset standard reaches the second preset duration, the cloud platform big data analysis module (2) sends a water tank self-cleaning equipment starting instruction to the water tank self-cleaning equipment in the water storage tank (4) corresponding to the water supply of the user in the pump room (3), and the water tank self-cleaning equipment is started to clean and purify the water in the water storage tank (4). And the first preset time length is less than the second preset time length.
Wherein, the above-mentioned real-time quality of water data is the digital expression of the real-time quality of water condition in corresponding water storage box (4), the above-mentioned step of confirming whether the quality of water in corresponding water storage box (4) of this pump house reaches the preset standard according to the real-time quality of water data in this real-time data includes:
and judging whether the real-time water quality data is smaller than the preset water quality data, if so, determining that the water quality in the corresponding water storage tank (4) of the pump room (3) reaches the preset standard, and if not, determining that the water quality in the corresponding water storage tank (4) of the pump room (3) does not reach the preset standard.
When the duration that the water quality in the corresponding water storage tank (4) of the pump room (3) does not reach the preset standard reaches a first preset duration, the water quality is short and exceeds the standard; at the moment, the cloud platform big data analysis module (2) sends a sterilizer opening instruction to a sterilizer in the water storage tank (4) corresponding to water supply of a user in the pump room (3), and the sterilizer is opened to sterilize the water in the water storage tank (4); when the duration that the water quality does not reach the preset standard reaches a second preset duration, the water quality exceeds the standard for a long time and is seriously polluted, at the moment, the cloud platform big data analysis module (2) sends a water tank self-cleaning equipment starting instruction to the water tank self-cleaning equipment in the water storage tank (4) corresponding to the water supply of the user in the pump room (3), and the water tank self-cleaning equipment is started to clean and purify the water body in the water storage tank (4). Therefore, by arranging the disinfector and the self-cleaning equipment in the cloud platform big data analysis module (2) to treat the water in the water storage tank (4), when the water quality in the water storage tank (4) does not reach the preset standard, the disinfector is controlled to disinfect the water in the water storage tank (4) or the self-cleaning equipment is controlled to clean the water in the water storage tank (4), so that the water quality safety of water supplied by a user using the water supply system is ensured.
In conclusion, the energy-saving peak shifting intelligent water supply system based on the cloud technology can control each intelligent water tank regulation and control unit to close to store water into the water storage tank (4) when the real-time water storage amount reaches the target water storage amount through the cloud platform technology; in the cloud platform technology integrated management area, the regulation and storage capacity of all the water storage tanks (4) realizes the peak-shifting regulation and storage water supply to the water supply pipe network system, and the pressure fluctuation range of the municipal water supply pipe network and the water supply pipe assembly is shortened by the peak-shifting regulation and storage water supply; the operation of the disinfector and the self-cleaning equipment corresponding to the water storage tank (4) of each pump room (3) in the management area is planned through the cloud platform technology, so that when the water quality in the water storage tank (4) does not reach a preset standard, the corresponding disinfector is controlled to disinfect the water body in the water storage tank (4) or the self-cleaning equipment is controlled to clean and purify the water body in the water storage tank (4), and the water quality safety of water supplied by a user using the water supply system is ensured. The invention can save national water supply resources and supporting facilities and ensure the water quality safety of water supply for users.
Those of ordinary skill in the art will understand that: the above-described figures are merely schematic representations of one embodiment, and the blocks or flows in the figures are not necessarily required to achieve all of the advantages of the present invention; the modules in the configuration device of the present invention can be collectively distributed in one device of the present invention according to all the contents described in the embodiments of the present invention, and can also be correspondingly changed to be respectively centralized in two or more devices of the present invention. That is, the modules in the embodiments of the present invention can be combined into one module, and can also be further split into multiple sub-modules.
The above embodiments are only used to illustrate the implementation schemes of the specific technologies of the present invention, and do not limit the protection scope; the present invention has been described in detail with reference to the above embodiments only, so that those skilled in the art can understand and realize the invention; other modifications to the technical solutions described in the above embodiments, or equivalent substitutions for some technical features, are possible without departing from the spirit and scope of the present invention, and are within the scope of the present invention.

Claims (6)

1. The energy-saving peak shifting intelligent water supply system based on the cloud technology comprises a pump room, water storage tanks, a plurality of water pump units and water supply pipeline assemblies, wherein the water storage tanks are installed in the pump room and are connected with the water pump units through the water supply pipeline assemblies, the water storage tanks of the pump room are connected with the water pump units through the water supply pipeline assemblies, the water pump units are installed in each pump room and are provided with water storage tanks, each water storage tank is connected with a water supply user of each water pump unit through the water supply pipeline assemblies, the water storage tanks in the pump rooms are connected with the water pump units through the water supply pipeline assemblies, and the pump rooms are connected through a municipal water supply pipeline network The hydraulic data monitoring unit, the water quality monitoring and processing unit, the pipe network hydraulic data monitoring unit, the pump room, the water storage tank, the water pump unit and the water supply pipeline assembly are connected with each other, the intelligent water tank regulating and controlling unit hydraulic data monitoring unit water quality monitoring and processing unit pipe network hydraulic data monitoring unit can transmit hydraulic water quality information data of the pump room, the water storage tank, the water pump unit and the water supply pipeline assembly which are collected in real time to the cloud platform server through the pump room monitoring unit and the data collecting and transmitting unit, and instruct the cloud platform big data analysis module according to the cloud platform server to regulate the pump room, the water storage tank, the water pump unit and the water supply pipeline assembly which exceed preset standard hydraulic water quality information data so that the hydraulic water quality information data which exceed the preset standard are regulated to be in accordance with the standard, and the cloud platform big data analysis module is embedded in the cloud platform server, the intelligent water tank water pump system comprises a data acquisition and transmission unit, a pump room monitoring unit, a smart water tank regulation and control unit, a hydraulic data monitoring unit, a water quality monitoring and processing unit and a pipe network hydraulic data monitoring unit, wherein the number of the data acquisition and transmission unit, the pump room monitoring unit, the water quality monitoring and processing unit and the pipe network hydraulic data monitoring unit is equal to that of a plurality of pump room water tank water pump units, the pump room monitoring unit of each pump room can be used for sending and transmitting monitoring data of the smart water tank regulation and control unit, the hydraulic data monitoring unit, the water quality monitoring and processing unit and the pipe network hydraulic data monitoring unit in the pump room, the pump room monitoring unit of each pump room sends real-time data to a cloud platform big data analysis module through the data acquisition and transmission unit of the pump room, the cloud platform big data analysis module determines target water storage amount of a corresponding water tank of the real-time data based on a pre-trained target neural network, The real-time water supply data and the hydraulic data at the tail end of the pipe network monitored by the intelligent water tank regulating and controlling unit and the pipe network hydraulic data monitoring unit determine whether to store water in the water storage tank;
the pump room monitoring unit of each pump room can be used for monitoring the pressure, flow, real-time water supply quantity and real-time water consumption data of the user, which are sent by the hydraulic data monitoring unit of the pump room, for supplying water to the user;
the pump room monitoring unit of each pump room can be used for monitoring the hydraulic data of the pressure, the flow, the real-time water supply amount and the real-time water supply amount data of the water supplied to the pump room water storage tank by the pipeline sent by the pipe network hydraulic data monitoring unit of the pump room;
the pump room monitoring unit of each pump room can be used for controlling the water quality monitoring and processing unit, the water quality analyzing equipment and the water quality sterilizing equipment of the pump room;
the pump room monitoring unit of each pump room can be used for analyzing the water quality of the water storage tank by the intelligent water tank regulating and controlling unit of the pump room, monitoring and controlling the water quality of the water storage tank in real time, transmitting the water quality data information monitored in real time to the cloud platform server and the cloud platform big data analysis module through the pump room monitoring unit and the data acquisition and transmission unit, and instructing the cloud platform big data analysis module to disinfect the water body with the water quality exceeding the preset standard according to the cloud platform server;
the cloud technology used by the cloud platform server and the cloud platform big data analysis module comprises public cloud, private cloud and mixed cloud, the cloud technology can store system data in a distributed storage mode, a redundant storage mode and a cold and hot backup storage mode, and the cloud technology enables a cloud system to construct distributed computing, utility computing, load balancing computing and parallel computing through virtualization basic hardware resources;
the energy-saving peak-shifting intelligent water supply system based on the cloud technology can be connected with a pump room, a water storage tank, a water pump unit, a water supply pipeline assembly, a computer and matched equipment in the pump room through a local area network technology to perform data communication and transmit instructions by constructing a unified platform formed by a plurality of cloud platform servers or storage server hardware;
the cloud technology-based energy-saving peak shifting intelligent water supply system can construct a public network-based local area network by adopting an IPsec VPN virtual private network technology based on public communication network infrastructure, and realize data transmission between a cloud end and matched equipment in each pump room;
the energy-saving peak shifting intelligent water supply system based on the cloud technology uses a PLC as a pump room field controller to acquire field data and receive a control instruction, and performs data communication with a cloud end through a Modbus industrial bus protocol standard;
the storage of pump room data in the energy-saving peak-shifting intelligent water supply system based on the cloud technology is realized by a distributed database, and the file processing is realized by processing of a distributed file system;
the energy-saving peak shifting smart water supply system based on the cloud technology acquires pump room data and completes cloud storage by constructing a local area network compatible with a public communication network at the cloud end, and a cloud platform big data analysis module is applied to a cloud platform server end to analyze water use data of each pump room so as to send a regulation and control instruction;
the cloud platform big data analysis module is used for determining the real-time water consumption of a user of the intelligent water tank regulation and control unit and the real-time data sent by the data acquisition and transmission unit of each pump room, determining the target water storage amount of the water storage tank corresponding to the real-time data based on a pre-trained target neural network model for the real-time data of each pump room, sending a starting instruction to the intelligent water tank regulation and control unit when the real-time water storage amount of the corresponding water storage tank in the real-time data does not reach the target water storage amount and the intelligent water tank regulation and control unit is not opened, so that the intelligent water tank regulation and control unit starts to store water for the water storage tank, calculating a first target water storage time based on the real-time water storage amount in the real-time data, the highest water storage amount of the water storage tank sent by the intelligent water tank regulation and control unit and the target water storage amount, and sending a first stopping instruction to the intelligent water tank, the intelligent water tank regulating and controlling unit is closed, when the real-time water storage amount in the real-time data does not reach the target water storage amount and the intelligent water tank regulating and controlling unit is opened, a second target water storage time is calculated based on the real-time water storage amount in the real-time data, the rated water supply amount of the intelligent water tank regulating and controlling unit and the target water storage amount, when the second target water storage time is reached, a second stop instruction is sent to the intelligent water tank regulating and controlling unit so that the intelligent water tank regulating and controlling unit stops storing water to the water storage tank, when the real-time water storage amount in the real-time data reaches the target water storage amount and the intelligent water tank regulating and controlling unit is opened for storing water, a third stop instruction is sent to the intelligent water tank regulating and controlling unit so that the intelligent water tank regulating and controlling unit stops storing water to the water storage tank, the target neural network model is based on real-time water consumption of a user, real-time water consumption of the user and sample water, training an initial target neural network model to obtain an initial neural network model, wherein the target neural network model can enable user real-time water consumption sample data to be associated with sample water storage amount of a corresponding water storage tank, and the user real-time water consumption sample data comprises the sample real-time water storage amount of the corresponding water storage tank, the sample real-time user water consumption, sample accumulated water supply amount, sample real-time water quality data and sample real-time water supply amount;
the cloud platform big data analysis module can be further used for performing discrete cosine transform on the real-time data based on a discrete cosine transform matrix to obtain redundancy-removed real-time data before determining the target water storage capacity of the water storage tank corresponding to the real-time data based on a pre-trained target neural network model, calculating a mean value and a standard deviation of the redundancy-removed real-time data, normalizing the redundancy-removed real-time data based on the mean value and the standard deviation to obtain normalized real-time data, and determining the target water storage capacity of the water storage tank corresponding to the normalized real-time data based on the pre-trained target neural network model;
the cloud platform big data analysis module can be specifically used for arranging real-time water storage amount, real-time user water consumption, accumulated water supply amount, real-time water quality data and sampling point data contained in the real-time water supply amount of a corresponding water storage tank in real-time data into a first matrix, calculating a product of the first matrix and a discrete cosine transform matrix to obtain a second matrix, and taking each element in the second matrix as the real-time data corresponding to the pump room after redundancy removal.
2. The cloud-technology-based energy-saving peak shifting intelligent water supply system according to claim 1, wherein the cloud platform big data analysis module is further configured to calculate a first mean value and a first standard deviation of first elements representing real-time water storage capacity in the second matrix, calculate, for each first element, a first difference value between the first element and the corresponding first mean value, and calculate a first quotient value between the first difference value and the first standard deviation;
calculating a second mean value and a second standard deviation of second elements representing the water consumption of real-time users in the second matrix, calculating a second difference value of each second element and the corresponding second mean value, and calculating a second quotient of the second difference value and the second standard deviation;
calculating a third mean value and a third standard deviation of third elements representing accumulated water supply in the second matrix, calculating a third difference value of each third element and the third mean value corresponding to the third element, and calculating a third quotient of the third difference value and the third standard deviation;
calculating a fourth mean value and a fourth standard deviation of a fourth element representing real-time water quality data in the second matrix, calculating a fourth difference value of each fourth element and the corresponding fourth mean value, and calculating a fourth quotient of the fourth difference value and the fourth standard deviation;
calculating a fifth mean value and a fifth standard deviation of fifth elements representing real-time water supply in the second matrix, calculating a fifth difference value of each fifth element and the corresponding fifth mean value, and calculating a fifth quotient value of the fifth difference value and the fifth standard deviation;
and taking the first quotient value, the second quotient value, the third quotient value, the fourth quotient value and the fifth quotient value as the normalized real-time data.
3. The cloud-technology-based energy-saving peak-shifting intelligent water supply system according to claim 1, wherein the training process of the target neural network model specifically comprises:
acquiring user real-time water consumption sample data used for training and sample water storage amount of a corresponding water storage tank, wherein the user real-time water consumption sample data comprises the sample real-time water storage amount of the corresponding water storage tank, the sample real-time user water consumption, the sample accumulated water supply amount, the sample real-time water quality data and the sample real-time water supply amount;
inputting the real-time water consumption sample data of the user and the sample water storage amount of the corresponding water storage tank into an initial neural network model, wherein the initial neural network model comprises an input layer, a hidden layer, an output layer and an equation set relation established between the input layer and the output layer, and the equation set relation established between the input layer and the output layer is an excitation function;
according to the user real-time water consumption sample data and the sample water storage amount of a corresponding water storage tank, the weight coefficient between each node of the input layer and each node of the hidden layer, the weight coefficient between each node of the hidden layer and each node of the output layer and the equation set relation established between the input layer and the output layer;
solving the equation set to obtain the values of the weight coefficients between each node of the input layer and each node of the hidden layer and the values of the weight coefficients between each node of the hidden layer and the nodes of the output layer;
substituting the values of the weight coefficients between the nodes of the input layer and the nodes of the hidden layer, the values of the weight coefficients between the nodes of the hidden layer and the nodes of the output layer and the user real-time water consumption sample data into the equation set to solve the initial water storage capacity of the water storage tank corresponding to the user real-time water consumption sample data;
calculating a difference value between the initial water storage amount of the corresponding water storage tank and the sample water storage amount of the corresponding water storage tank;
when the difference value is larger than a preset difference threshold value, adjusting a first weight coefficient and a second weight coefficient based on the difference value, and returning to the step of inputting the user real-time water consumption sample data and the sample water storage amount of the corresponding water storage tank into an initial neural network model, wherein the first weight coefficient is a first weight coefficient between a first input layer node and a first hidden layer node, and the second weight coefficient is a second weight coefficient between a second input layer node and the first hidden layer node;
when the difference value is not greater than a preset difference threshold value, finishing training to obtain a target neural network model containing the corresponding relation between the real-time water consumption sample data of the user and the sample water storage capacity of the corresponding water storage tank;
the input layer comprises a first input layer node, a second input layer node and a third input layer node, the hidden layer comprises a first hidden layer node, a second hidden layer node and a third hidden layer node, the output layer comprises an output layer node, and the step of establishing the equation set relationship between the input layer and the output layer according to the user real-time water consumption sample data and the sample water storage capacity of the corresponding water storage tank, the weight coefficient between each node of the input layer and each node of the hidden layer, the weight coefficient between each node of the hidden layer and the output layer node of the output layer and the equation set relationship between the input layer and the output layer comprises the following steps:
establishing a first linear equation according to the user real-time water consumption sample data, a first weight coefficient between the first input layer node and the first hidden layer node, a second weight coefficient between the second input layer node and the first hidden layer node and a first constant between the third input layer node and the first hidden layer node;
establishing a second linear equation according to the user real-time water consumption sample data, a third weight coefficient between the first input layer node and the second hidden layer node, a fourth weight coefficient between the second input layer node and the second hidden layer node, and a second constant between the third input layer node and the second hidden layer node;
establishing a third linear equation according to the user real-time water consumption sample data, a fifth weight coefficient between the first input layer node and the third hidden layer node, a sixth weight coefficient between the second input layer node and the third hidden layer node, and a third constant between the third input layer node and the third hidden layer node;
establishing a fourth function equation according to the equation set relation established between the input layer and the output layer, wherein the equation set relation is an excitation function, the sample water storage capacity of a water storage tank corresponding to the user real-time water consumption sample data, the function value of the first linear equation, the function value of the second linear equation and the function value of the third linear equation, a seventh weight coefficient between the first hidden layer node and the output layer node, an eighth weight coefficient between the second hidden layer node and the output layer node and a ninth weight coefficient between the third hidden layer node and the output layer node;
the first linear equation, the second linear equation, the third linear equation and the fourth function equation are simultaneously established to obtain an equation set between the input layer and the output layer;
the step of solving the equation set to obtain the values of the weight coefficients between the nodes of the input layer and the nodes of the hidden layer and the values of the weight coefficients between the nodes of the hidden layer and the nodes of the output layer includes:
solving the equation set to obtain a value of the first weight, a value of the second weight, a value of the third weight, a value of the fourth weight, a value of the fifth weight, a value of the sixth weight, a value of the seventh weight, a value of the eighth weight, and a value of the ninth weight;
substituting the values of the weight coefficients between the nodes of the input layer and the nodes of the hidden layer, the values of the weight coefficients between the nodes of the hidden layer and the nodes of the output layer and the user real-time water consumption sample data into the equation set to solve the initial water storage capacity of the water storage tank corresponding to the user real-time water consumption sample data, wherein the step comprises the following steps:
and substituting the value of the first weight, the value of the second weight, the value of the third weight, the value of the fourth weight, the value of the fifth weight, the value of the sixth weight, the value of the seventh weight, the value of the eighth weight, the value of the ninth weight and the user real-time water consumption sample data into the equation set to solve the initial water storage capacity of the water storage tank corresponding to the user real-time water consumption sample data.
4. The cloud technology-based energy-saving peak-shifting intelligent water supply system according to claim 3, wherein the function value of the excitation function is: the sum of the excitation coefficient and a target maximum value, wherein the target maximum value is the maximum value in the user real-time water consumption sample data and a target product, and the target product is the product of the excitation coefficient and the user real-time water consumption sample data.
5. The energy-saving peak shifting intelligent water supply system based on the cloud technology as claimed in claim 1, it is characterized in that a sterilizer and a water storage tank self-cleaning device are also arranged in the water storage tank of each pump room, the cloud platform big data analysis module can also be used for determining whether the water quality in the water storage tank of the pump room reaches a preset standard according to the real-time water quality data in the real-time data, when the duration of the water quality which does not reach the preset standard reaches the first preset duration, sending a sterilizer opening instruction to the sterilizer in the pump room water storage tank to open the sterilizer, when the duration of the water quality which does not reach the preset standard reaches a second preset duration, sending a water storage tank self-cleaning equipment starting instruction to water storage tank self-cleaning equipment in the pump room water storage tank, so that the water storage tank self-cleaning equipment is started, and the first preset time is shorter than the second preset time.
6. The cloud-technology-based energy-saving peak shifting intelligent water supply system according to claim 5, wherein the real-time water quality data is real-time water quality, and the step of determining whether the water quality in the water storage tank of the pump room meets a preset standard according to the real-time water quality data in the real-time data comprises:
and judging whether the real-time water quality is less than the preset water quality, if so, determining that the water quality in the water storage tank of the pump room reaches the preset standard, and if not, determining that the water quality of the water stored in the pump room does not reach the preset standard.
CN201910612764.9A 2019-07-09 2019-07-09 Energy-saving peak shifting intelligent water supply system based on cloud technology Active CN110284556B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910612764.9A CN110284556B (en) 2019-07-09 2019-07-09 Energy-saving peak shifting intelligent water supply system based on cloud technology

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910612764.9A CN110284556B (en) 2019-07-09 2019-07-09 Energy-saving peak shifting intelligent water supply system based on cloud technology

Publications (2)

Publication Number Publication Date
CN110284556A CN110284556A (en) 2019-09-27
CN110284556B true CN110284556B (en) 2020-04-21

Family

ID=68022009

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910612764.9A Active CN110284556B (en) 2019-07-09 2019-07-09 Energy-saving peak shifting intelligent water supply system based on cloud technology

Country Status (1)

Country Link
CN (1) CN110284556B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110764481B (en) * 2019-11-10 2022-06-07 许刚 Distributed redundant constant-pressure water supply control system
CN110965610B (en) * 2019-11-25 2021-10-22 熊猫智慧水务有限公司 Double-circuit peak regulation equipment based on MPC control
CN111680429B (en) * 2020-06-19 2020-12-22 苏州华控清源系统科技股份有限公司 Water tank active storage adjusting method and system, electronic equipment and storage medium
CN112482488B (en) * 2020-11-11 2022-11-22 上海威派格智慧水务股份有限公司 Urban peak shifting water supply method and system
CN112783125A (en) * 2020-12-31 2021-05-11 深圳市水务技术服务有限公司 Intelligent water supply information acquisition and scheduling system and method
CN112881630A (en) * 2021-01-18 2021-06-01 青岛黄海学院 Water quality monitoring system based on cooperative communication
CN112696616B (en) * 2021-01-29 2022-08-05 浙江和达科技股份有限公司 Water supply network wisdom management and control system
CN113359552B (en) * 2021-06-03 2022-01-04 湖南柒丰智能科技有限公司 Water supply pump room automatic control system and method based on big data analysis
CN114637269B (en) * 2022-04-15 2023-04-07 安徽中科大国祯信息科技有限责任公司 Online control and scheduling system and method for water resources of smart park
CN117156313B (en) * 2023-08-29 2024-04-02 广东祁越智能科技有限公司 Intelligent pump house operation safety detection system based on Internet of things

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN205189042U (en) * 2015-11-17 2016-04-27 安徽舜禹水务股份有限公司 Novel can set for water supply system that avoids peak hour
CN105464171A (en) * 2016-01-04 2016-04-06 安徽舜禹水务股份有限公司 Intelligent peak staggering capacity regulating security water supply system
CN205276378U (en) * 2016-01-04 2016-06-01 安徽舜禹水务股份有限公司 Intelligence regulation security protection water supply system that avoids peak hour
WO2018102939A1 (en) * 2016-12-05 2018-06-14 王彩霞 Prefabricated pump station for tap water pressurization
CN108755842A (en) * 2018-05-23 2018-11-06 成都优链加科技有限公司 A kind of secondary water supply system is avoided the peak hour moisturizing management and running platform

Also Published As

Publication number Publication date
CN110284556A (en) 2019-09-27

Similar Documents

Publication Publication Date Title
CN110284556B (en) Energy-saving peak shifting intelligent water supply system based on cloud technology
CN102629106B (en) Water supply control method and water supply control system
CN107274303A (en) City heat supply energy consumption wisdom management and control technology platform
CN110410942A (en) A kind of Cooling and Heat Source machine room energy-saving optimal control method and system
CN103453580B (en) Heat exchange station integrated intelligent control method
CN108279632B (en) A kind of pumping plant wisdom draining Dispatching Control System
CN108490772A (en) A kind of urban water affairs system steadily supplies water progress control method
CN106996623B (en) Central air conditioner room control method under energy consumption limit
CN116246388A (en) Device and method for apportioning measurement according to matching coefficient
CN112213953A (en) Intelligent building equipment control method, platform, equipment and computer storage medium
CN105373000B (en) Intelligence community platform control system and control method based on " internet+"
CN110848895B (en) Non-industrial air conditioner flexible load control method and system
CN115619136A (en) Building management method and system
Zhao et al. A systematic mixed-integer differential evolution approach for water network operational optimization
CN111160384B (en) Mobile energy storage vehicle scheduling optimization method, system and device
CN109523889B (en) Building secondary water supply energy-saving pilot test system and method
CN113309173B (en) Water supply system optimization method and adjustment method for community water supply system
CN205595796U (en) Electric wire netting demand response control system
CN111626603B (en) Oil field ordered electricity utilization system with genetic algorithm and ordered electricity utilization distribution method thereof
CN110826776B (en) Initial solution optimization method based on dynamic programming in distribution network line transformation relation identification
CN113849005A (en) Intelligent feedback type pipe network pressure control system and method
CN103825301B (en) A kind of unified interface modeling method of distributed energy access active distribution system
Pinanggih et al. Prediction of energy consumption using artificial neural network method in one of shopping center in Cirebon city
CN111639796B (en) Ordered oil field electricity utilization system with topological structure and ordered electricity utilization distribution method thereof
CN112600204B (en) Electric load prediction method and system for commercial integrated air conditioning equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CP02 Change in the address of a patent holder

Address after: 230000 No.8 Huainan North Road, Shuangfeng Economic Development Zone, Changfeng County, Hefei City, Anhui Province

Patentee after: ANHUI SHUNYU WATER AFFAIRS Co.,Ltd.

Address before: 230000 Jin Jiang Road, Changfeng Shuangfeng Economic Development Zone, Hefei, Anhui, 32

Patentee before: ANHUI SHUNYU WATER AFFAIRS Co.,Ltd.

CP02 Change in the address of a patent holder