CN109270016B - Multifunctional underwater water quality automatic monitoring system and monitoring method based on clustering algorithm - Google Patents

Abstract

The invention discloses a multifunctional underwater water quality automatic monitoring system and a monitoring method based on a clustering algorithm. When the monitor works, different reagents are used for titrating the current water sample, and the ultraviolet light source is controlled to emit different lights to monitor different water quality parameters; and meanwhile, the lower computer controls the sliding module to enable the box body to move up and down in the vertical direction, so that a plurality of water quality parameters of different water depths are collected. The system can measure COD, nitrogen content and PH value in different water depths, and upload data to the upper computer through the data access module for data processing based on the clustering algorithm, thereby realizing multifunction and automation of water quality monitoring, avoiding secondary pollution, saving cost, improving water quality monitoring accuracy and data processing and analyzing comprehensiveness.

Description

Multifunctional underwater water quality automatic monitoring system and monitoring method based on clustering algorithm

technical field

The invention relates to a water quality monitoring device and method, in particular to a multifunctional underwater water quality automatic monitoring system and monitoring method based on a clustering algorithm.

Background technique

With the development of industry, water pollution has become more and more serious. Chemical oxygen demand (COD), nitrogen content and PH value can reflect the degree of water pollution by reducing substances, which is an important standard for measuring environmental quality. At present, water quality The traditional method for the determination of water quality generally adopts chemical methods, mainly using potassium dichromate standard method and coulometric titration method. , seawater using alkaline potassium permanganate method. However, these traditional methods all have the disadvantages of large amount of reagents, serious secondary pollution, long test time, complicated operation steps, single measurement object, and low degree of automation, which are not suitable for the modern market demand.

The spectrophotometric method of water quality measurement is carried out on the basis of titration measurement, using the principle of absorption spectrum, according to the ion content of the reagent after the titration in the solution is completed, using ultraviolet light sources of different wavelengths to measure the water quality of the solution to be tested with different concentrations. Therefore, according to the obtained absorbance working curve, the chemical oxygen demand, nitrogen content and PH value in the water sample can be expressed. Spectrophotometry is an improvement of titration, which saves reagents and is easy to operate. However, most of the existing instruments can only measure one parameter of water quality, and cannot complete the unified measurement of multiple parameters, which also leads to one-sided and inaccurate water quality measurement results. Water quality monitoring equipment is imminent.

The existing spectrophotometry-based water quality monitors generally need to be equipped with special reagents from the corresponding production companies, and must require artificial titration reagents, manual sampling, and manual operations to complete various water quality monitoring steps, so a lot of human, material and financial resources are wasted. Moreover, the obtained water samples are single and cannot reflect the water quality in different depths of water environment. The monitoring results are not ideal, and the compatibility of water quality monitoring instruments in different measurement environments is very poor. The actual measurement of the water environment is one-sided, and the measurement speed and The accuracy needs to be improved, and the instrument is expensive and bulky, and the operation steps are complicated and inconvenient for practical application. Therefore, how to realize the integration of multi-function, low price, dynamic data access and offline processing, on-site measurement and remote monitoring, manual measurement and automatic measurement, water quality measurement at different water depths, simple and accurate measurement, fast and portable is the face of water quality monitoring instruments. one of the pressing issues.

SUMMARY OF THE INVENTION

Purpose of the invention: One purpose is to provide a multifunctional underwater water quality automatic monitoring system and monitoring method based on a clustering algorithm, so as to solve the above-mentioned problems existing in the existing technology and application, expand the function of the water quality monitor, and improve the water quality monitor for different water depths. Automatic measurement function and field measurement function of multiple parameters.

Technical scheme: In order to realize the above-mentioned purpose of the invention, the present invention adopts the following technical scheme:

A multifunctional underwater water quality automatic monitoring system based on a clustering algorithm, including a host computer and a monitor, the host computer includes a Labview interface design part and a Python data processing part, the host computer designs a GUI interface through Labview, and calls Python with the help of internal interface functions The function uses the clustering algorithm to complete the sample data processing. The monitor includes a retractable fixed bracket, a box body, a lower computer, a sliding module and a power supply module. in:

The lower end of the retractable fixing bracket is fixed in the water to fix the entire monitor;

The box body is sealed, a hole is opened at the top, a measuring switch is placed outside the box body, a waterproof layer is arranged on the inner wall of the box body, and the lower computer and the power module are located in the waterproof layer;

The lower computer is built in the box and is used to control the sliding module and collect water quality parameter data, and upload the collected water quality parameter data to the upper computer;

The sliding module is used to connect the box body and the retractable fixing bracket;

The power module is built in the box, including a DC power drive module and a voltage amplifier module. The output of the DC power drive module is connected to the input of the voltage amplifier module, and the output end of the voltage amplifier module is respectively connected to the power input end of each module to supply power to each module of the monitor;

The lower computer controls the sliding module, so that the box body moves up and down in the vertical direction, and then collects multiple water quality parameters of different water depths.

Optionally, the telescopic fixing bracket includes two upper and lower sections of stainless steel iron pipes, the upper section is a telescopic section, the lower section is a fixed section, the lower end of the telescopic section is provided with a spring latch, and the fixed section is distributed with a plurality of fixed sections that cooperate with the spring latches. When the spring bolts are inserted into different fixing holes, the entire telescopic fixing bracket has different heights, and the telescopic section and the fixing section are fixedly connected through the spring bolts and the fixing holes.

Optionally, the lower computer includes a single-chip microcomputer, a multi-channel spectrophotometric water quality measurement module, and a data access module. The multi-channel spectrophotometric water quality measurement module converts the measured water quality information into electrical signals, which are then output to the single-chip microcomputer for processing, and the single-chip computer outputs the data after processing. To the data access module, the host computer reads data from the data access module.

Optionally, the single-chip microcomputer includes a control unit, an external RTC, an external crystal oscillator, a signal conditioning and amplifying circuit, and a multi-channel ADC conversion module, and the multi-channel spectrophotometric water quality measurement module includes a fixed base, a dual-channel spectrophotometer, a test tube fixture, and a silicon photocell sensor. 1. Silicon photocell sensor 2, ultraviolet light source 1, ultraviolet light source 2, titration tank and water valve, the dual-channel spectrophotometer is fixed on the fixed base, and the inner wall of the test tube fixing device is provided with an optical path on-off switch, which is used to hold the test tube. The test tube of the reagent is placed in the test tube fixing device, and the opening of the test tube is facing the opening on the top of the box body, and the water valve is set at the upper end of the test tube fixing device. Close, open the titration tank to titrate the water quality in the current test tube; both the external RTC and the external crystal oscillator are connected to the control unit, and the control unit controls the two ultraviolet light sources to emit ultraviolet light of different intensities and frequencies by outputting PWM waves, and the ultraviolet light passes through the spectrophotometer respectively. The internal channel is irradiated on the corresponding silicon photocell sensor, and then the optical signal is converted into an electrical signal by the silicon photocell sensor, and the electrical signal is amplified by the input signal conditioning and amplifying circuit, and finally processed by the multi-channel ADC conversion module input control unit.

Optionally, the data access module includes an SD card data dynamic access module and an EEPROM data cache module, the SD card data dynamic access module is bidirectionally connected to the control unit and the host computer respectively, and the EEPROM data cache module is connected to the control unit.

Optionally, the sliding module includes a stepping motor, a fixed pulley and a movable pulley, the fixed pulley is fixed on the upper end of the telescopic fixing bracket, and the movable pulley is fixed on the outer wall of the box body on the side close to the telescopic fixing bracket, and is connected with the telescopic fixing bracket. The stepper motor is fixed on the upper part of the box body, the traction wire drawn from the fixed pulley is sleeved on the output shaft of the stepper motor, and the single chip controls the stepper motor to rotate clockwise or counterclockwise, which drives the box body to move up and down in the vertical direction. , and the movable pulley moves up and down with the box body.

In another embodiment of the present invention, a monitoring method for a multifunctional underwater water quality automatic monitoring system based on a clustering algorithm includes the following steps:

(1) Debug the monitor before starting the measurement: adjust the length of the retractable fixing bracket to the height suitable for the current water level, insert the retractable fixing bracket into the water, and insert it firmly;

(2) Judging whether the measuring switch is turned on, if the measuring switch is turned on, input the current water depth and the unit height value of the box body each time to the control unit, and execute step (3); if not, continue to execute step (2) ;

(3) Before starting monitoring, judge whether the current descended height is greater than or equal to the input water depth value. If it is found that the current descended height is greater than or equal to the input water depth value before starting monitoring, it means that the input in step (2) is wrong and needs to return Step (2) re-input the current water depth and the unit height value of the box body falling each time; if not, then execute step (4);

(4) The control unit controls the forward rotation of the stepper motor, so that the box body drops by a unit height value, and monitors different water quality parameter values at the current water depth position;

(5) After one measurement is over, the control unit recalculates and judges whether the current descended height is greater than or equal to the input water depth value. If not, it means that the instrument has not reached the bottom, and returns to step (4) to execute; if so, execute step (6) );

(6) The control unit controls the stepper motor to reverse, so that the box body rises by a unit height value;

(7) The control unit calculates and judges whether the current raised height is greater than or equal to the input water depth value. If not, it means that the instrument has not reached the water surface, and returns to step (6) to execute; if so, it means that the box body has reached the water surface. The control unit stores the measurement results in the data access module, and the measurement ends;

(8) After the measurement, connect the data access module to the host computer, Labview reads the data stored in the data access module through the internal function interface as the sample data set to be processed, and Labview calls the Python function through the internal interface function to use The clustering algorithm completes the sample data processing and displays it through the Labview design GUI interface.

Further, in step (4), the method for monitoring the multiple water quality parameter values of the current water depth position is:

(1) When the box body reaches the predetermined position, the control unit controls the opening of the water valve, so that the external water quality flows into the test tube along the water inlet above the box body. If the water level of the injected water reaches above the water valve, the control unit controls the water valve to close. , to achieve the purpose of taking water samples;

(2) The control unit controls the titration tank to titrate the current water sample with different reagents according to the water quality parameters that currently need to be measured, and closes the titration tank after the titration is completed;

(3) The control unit outputs a PWM wave with a frequency of i to the ultraviolet light source, so that it emits ultraviolet light of the corresponding frequency and intensity. The current water quality parameter α is measured by spectrophotometry, and the two silicon photocell sensors are used to measure N times respectively, and remove the maximum value. value and the minimum value, take the average value, and store the current water quality parameter results in the SD card respectively; where the water quality parameter α is the COD concentration, nitrogen content or PH value of the water quality, and i represents the current water quality parameter α required to obtain frequency of the PWM wave.

Further, in step (8), the host computer adopts a clustering algorithm to process the sample data, which specifically includes:

The system uses Labview to directly call the Python interface function to open the Python console, and uses the KMeans function in the built-in scikit-learn package in Python to encode the clustering algorithm program. When the interface is connected normally, after opening the Python console, you need to import the input through the data access module first. The water quality monitoring data sets of different positions and water depths in the upper computer part are taken as the sample data set D. According to the k value input by the user, k samples are randomly selected from D as the initial mean vector, and each sample data in the data set is calculated. The distance from each sub-vector in the mean vector, the distance metric function is calculated by using a Gaussian function: d _ij =||x _j -u _i || ₂ , 1≤i≤k, 1≤j≤m, where d _ij represents the Gaussian distance between the sample data and the mean vector, x _j represents the sample data to be divided, _ui represents the sample cluster formed by clustering, divide the sample x _j into the nearest cluster, and then recalculate the new The mean vector is determined and whether it is consistent with the previous mean vector. If it is consistent, the clustering is completed, and the sample data set is divided into k clusters. If it is inconsistent, the above calculation steps are repeated until the mean vector no longer changes.

Beneficial effects: compared with the prior art, the present invention adopts multi-channel spectrophotometry to measure COD, nitrogen content and pH value at different water depths, realizes instrument automation, avoids secondary pollution, saves costs, and improves the accuracy of water quality monitoring. At the same time, it saves a lot of manpower, material and financial resources. At the same time, the instrument supports on-site measurement and remote monitoring, especially suitable for field water quality measurement; the instrument uses multiple channels to measure different water quality parameters, effectively reducing irrelevant factors for water quality measurement. interference, making the measurement more accurate and fast; the introduction of the SD card data dynamic access function makes the operation easier and more convenient.

At the same time, the existing water quality monitoring instruments only complete a single data collection function, and there is not much involvement in the later data processing, and due to the numerous parameters of water quality monitoring, it is difficult to add appropriate labels to the sample data to evaluate the quality of water quality. , the clustering algorithm in the field of machine learning is suitable for the classification and data processing of unlabeled data sets. By learning the unlabeled training samples to explain the inherent nature and laws of the data, it can be used to find the inherent distribution structure of the data, or As a precursor to other learning tasks such as classification. At the same time, the instrument is small in size, easy to carry, and low in price, which is closer to practical applications. It meets the needs of industry and individuals to a certain extent, and helps to further improve the problem of environmental pollution. It has high social practical value and environmental protection value.

Description of drawings

Fig. 1 is the structural representation of monitor of the present invention;

Fig. 2 is the circuit connection schematic diagram of the monitor of the present invention;

Fig. 3 is the measurement flow chart of the present invention;

Fig. 4 is the method flow chart of measuring different water quality parameters of the present invention;

FIG. 5 is a schematic diagram of the clustering algorithm of the present invention.

Detailed ways

The technical solutions of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

The invention implements a multifunctional underwater water quality automatic monitoring system and monitoring method based on a clustering algorithm, which can realize multi-channel spectrophotometry to measure the COD concentration, nitrogen content and PH value of water quality at different water depths, and automatically complete water quality sampling at different water depths. , titration, parameter measurement, SD card data dynamic access, on-site measurement and offline monitoring and data processing functions. Among them, STM32F103RBT6 is the main chip (that is, the control unit, the same below), which coordinates the orderly work of each module.

A multifunctional underwater water quality automatic monitoring system based on a clustering algorithm, including a host computer and a monitor, the host computer includes a Labview interface design part and a Python data processing part, the host computer designs a GUI interface through Labview, and calls Python with the help of internal interface functions Complete sample data processing, the sample data is a data set composed of multiple measured water quality parameters read from the data access module to the host computer, and with the increase of new water quality parameters, this data set is dynamic Variable. The monitor includes a retractable fixed bracket, a box, a lower computer, a sliding module and a power module. The instrument measurement is automated, the measurement parameters are comprehensive, and the measurement results are simple and accurate, especially suitable for offline water quality monitoring and water quality monitoring at different water depths in the wild. Among them, the lower end of the retractable fixing bracket is fixed in the water to fix the entire monitor; the box body is sealed with a hole at the top, a measuring switch is placed outside the box body, the inner wall of the box is provided with a waterproof layer, and the lower computer and the power module It is located in the waterproof layer; the lower computer is built in the box, which is used to control the sliding module and collect water quality parameter data, and upload the collected water quality parameter data to the upper computer; the sliding module is used to connect the box body and the retractable fixing bracket; the power module It is built in the box and includes a DC power drive module and a voltage amplifier module. The output of the DC power drive module is connected to the input of the voltage amplifier module, and the output end of the voltage amplifier module is respectively connected to the power input end of each module to supply power to each module of the monitor.

As shown in Figures 1 and 2, the multifunctional underwater water quality automatic monitor based on the clustering algorithm, wherein the retractable fixed bracket is composed of two sections of stainless steel pipes with adjustable height, the length of which can be adjusted according to the water depth. The adjustable length of the retractable fixing bracket is long enough to ensure that the highest point exceeds the height of the water surface by at least one box body, so as to ensure that the box body is above the water surface in the initial position; the retractable fixing bracket is divided into upper and lower sections, the upper The section is a telescopic section, the lower section is a fixed section, the lower end of the telescopic section is provided with a spring latch, and the fixed section is distributed with a plurality of fixing holes at the same distance. It has different heights. When you need to adjust the height, you only need to press the spring latch into the fixing hole to move the telescopic joint up and down to adjust the height of the telescopic fixing bracket. When the box body needs to be lowered, it is only necessary to control the rotation of the stepper motor, and the box body will be lowered for a certain distance along the retractable fixed bracket through the fixed pulley and the movable pulley.

The box body only has an opening at the water inlet, the other four sides are sealed, and a waterproof layer is built in the box body to prevent the water ingress of the instrument from damaging the circuit. It is used to start and close the monitor; the box body is connected to the adjustable fixed bracket through the sliding module.

The lower computer includes a single-chip microcomputer, a multi-channel spectrophotometric water quality measurement module and a data access module. The single-chip microcomputer, as the core board of the instrument, is placed at the bottom of the box, and is respectively connected with other modules to complete data processing and command control. The single-chip microcomputer includes A control unit, an external RTC, an external crystal oscillator, a signal conditioning amplifier circuit, and a multi-channel ADC conversion module. The multi-channel spectrophotometric water quality measurement module includes a fixed base, a dual-channel spectrophotometer, a test tube fixture, a silicon photocell sensor 1, and a silicon photocell Sensor 2, UV light source 1, UV light source 2, titration tank and water valve, the dual-channel spectrophotometer is fixed on the fixed base, and there is an optical path on-off switch on the inner wall of the test tube fixing device, which is used for the test tube containing the test agent. In the test tube fixing device, and its opening is facing the water inlet of the box body, the water valve is set at the upper end of the test tube fixing device. When the water to be tested is filled to the point where the liquid level is above the water valve, the control unit controls the water valve to close and open the titration tank. Titrate the water quality in the current test tube; the external RTC and the external crystal oscillator are connected to the control unit, and the control unit controls the two ultraviolet light sources to emit ultraviolet light of different intensities and frequencies by outputting PWM waves. The two ends of the optical channel are respectively placed with an ultraviolet light source and a silicon photocell sensor. The ultraviolet light is irradiated to the corresponding silicon photocell sensor through the internal channel of the spectrophotometer, and then the optical signal is converted into an electrical signal through the silicon photocell sensor, and then input The signal conditioning and amplifying circuit amplifies the electrical signal, and finally it is processed by the multi-channel ADC conversion module input control unit; two ultraviolet light sources can be used to measure water quality parameters in different data ranges respectively, and different sufficient titration reagents are placed in the titration tank. , used to titrate the water quality stock solution to form the test agent, the water valve is directly connected to the pin of the control unit, and the control unit directly controls its opening and closing. The control unit adopts the ARM processing chip STM32F103RBT6, and the signal conditioning and amplifying circuit includes a voltage amplifying circuit composed of an OPA2336UA chip and a resistance-capacitance filter circuit for amplifying and filtering analog signals; the multi-channel ADC conversion module is equipped with AD7705 High-precision analog-to-digital conversion chip, the acquisition accuracy can reach 16 bits, using SPI mode to communicate with the control unit, and using DMA mode to achieve simultaneous acquisition of multi-channel signals.

The data access module includes an SD card data dynamic access module and an EEPROM data cache module, the SD card data dynamic access module is bidirectionally connected to the control unit and the host computer, and the EEPROM data cache module is connected to the control unit; complete data offline storage. The power module is built into the box, including a DC power drive module and a voltage amplifier module. The output of the DC power drive module is connected to the input of the voltage amplifier module, and the output terminals of the voltage amplifier module are respectively connected to the power input terminals of each module to supply power to each module of the instrument. . The voltage amplifying module is powered by a DC power supply driving module, and an amplifying circuit is designed according to a plurality of PT1301 chips to provide working voltages for different modules respectively. The DC power drive module consists of two 1.5v dry batteries.

The sliding module includes a stepping motor, a fixed pulley and a movable pulley. The fixed pulley is fixed on the top of the telescopic fixing bracket, and the movable pulley is fixed on the outer side wall of the box body near the telescopic fixing bracket, and is closely attached to the telescopic fixing bracket. The stepping motor is fixed on the upper part of the box body, and the traction wire drawn from the fixed pulley is sleeved on the output shaft of the stepping motor. If the stepping motor rotates clockwise or counterclockwise, it will drive the box body to move in the vertical direction, and the moving pulley It moves up and down in the vertical direction, and the movable pulley moves up and down with the box body, and plays the role of lubricating the movement of the box body and fixing the box body to the telescopic fixing bracket. Among them, the fixed end of the stepping motor is welded on the upper left of the box body, and the box body is connected with the retractable fixed bracket through the sliding module. By controlling the rotation of the stepping motor, the purpose of adjusting its own height and measuring the water quality of different water depths is achieved.

As shown in Figure 3, the overall monitoring method of the multifunctional underwater water quality automatic monitor based on the clustering algorithm is as follows:

(1) Debug the monitor before starting the measurement: adjust the length of the retractable fixing bracket to the height suitable for the current water level, insert the retractable fixing bracket into the water, and insert it firmly;

(2) Judging whether the measuring switch is turned on, if the measuring switch is turned on, input the current water depth and the unit height value of the box body each time to the control unit, and execute step (3); if not, continue to execute step (2) ;

(3) Before starting monitoring, judge whether the current descended height is greater than or equal to the input water depth value. If it is found that the current descended height is greater than or equal to the input water depth value before starting monitoring, it means that the input in step (2) is wrong and needs to return Step (2) re-input the current water depth and the unit height value of the box body falling each time; if not, then execute step (4);

(4) The control unit controls the forward rotation of the stepper motor, so that the box body drops by a unit height value, and monitors different water quality parameter values at the current water depth position;

(5) After one measurement is over, the control unit recalculates and judges whether the current descended height is greater than or equal to the input water depth value. If not, it means that the instrument has not reached the bottom, and returns to step (4) to execute; if so, execute step (6) );

(6) The control unit controls the stepper motor to reverse, so that the box body rises by a unit height value;

(7) The control unit calculates and judges whether the current raised height is greater than or equal to the input water depth value. If not, it means that the instrument has not reached the water surface, and returns to step (6) to execute; if so, it means that the box body has reached the water surface. This measurement is over;

(8) After the measurement, connect the data access module to the host computer, Labview reads the data stored in the data access module through the internal function interface as the sample data set to be processed, and Labview calls the Python function through the internal interface function to use The clustering algorithm completes the sample data processing and displays it through the Labview design GUI interface.

Referring to Figure 4, the concrete methods and steps of measuring different water quality parameters in the step (4) are:

(41) When the box body reaches the predetermined position, the control unit controls the opening of the water valve, so that the external water quality flows into the test tube along the water inlet above the box body. If the water level of the injected water reaches above the water valve, the control unit controls the water valve to close. , to achieve the purpose of taking water samples;

(42) the control unit controls the titration tank to titrate the current water sample with different reagents according to the water quality parameters currently required to be measured, and closes the titration tank after the titration is completed;

(43) The control unit outputs a PWM wave with a frequency of i to the ultraviolet light source, so that it emits ultraviolet light of the corresponding frequency and intensity, uses the spectrophotometry to measure the current water quality parameter α, and uses two silicon photocell sensors to measure N times respectively, remove the maximum value and the minimum value, take the average value, and store the current water quality parameter results in the SD card respectively; where the water quality parameter α is the COD concentration, nitrogen content or PH value of the water quality, and i represents the current water quality parameter α required to obtain frequency of the PWM wave.

Referring to Figure 5, the clustering algorithm is a common unsupervised learning algorithm in the field of machine learning. It is suitable for cases where the labeling of training samples is unknown. The goal is to reveal the inherent nature and laws of sample data through the learning of unlabeled training samples. , to prepare for further data analysis, the essence of which is to divide the sample data set into several disjoint subsets, that is, sample clusters. The clustering results are judged by the similarity within clusters and between clusters. In this system, the k-means algorithm is used to complete the clustering, and the sample data is divided into several clusters, and the classification results are displayed through the Labview user interface. There are k sample clusters. Users can divide the water quality at different locations and depths according to the generated sample clusters, and judge different water pollution conditions differently. They can also analyze the sample clusters in areas with poor water quality or water depth to find this The main influencing factors of regional water pollution are which parameters of water quality (COD, PH, nitrogen content, etc.), and which depth of water generally exists in areas with more serious water pollution. The k value is input and specified by the user.

In the above step (8), the host computer uses a clustering algorithm to process the sample data, which specifically includes:

The system uses Labview to directly call the Python interface function to open the Python console, and uses the KMeans function in the built-in scikit-learn package in Python to encode the clustering algorithm program. When the interface is connected normally, after opening the Python console, you need to import the input through the data access module first. The water quality monitoring data sets of different positions and water depths in the upper computer part are taken as the sample data set D. According to the k value input by the user, k samples are randomly selected from D as the initial mean vector, and each sample data in the data set is calculated. The distance from each sub-vector in the mean vector, the distance metric function is calculated by using a Gaussian function: d _ij =||x _j -u _i || ₂ , 1≤i≤k, 1≤j≤m, where d _ij represents the Gaussian distance between the sample data and the mean vector, x _j represents the sample data to be divided, _ui represents the sample cluster formed by clustering, divide the sample x _j into the nearest cluster, and then recalculate the new mean vector, and judge whether it is consistent with the previous mean vector. If it is consistent, the clustering is completed, and the sample data set is divided into k clusters. If it is inconsistent, the above calculation steps are repeated until the mean vector does not change. In actual operation Using the KMeans function directly can replace the above calculation steps to complete the clustering task.

In a word, the present invention adopts STM32 to realize multi-channel spectrophotometric measurement of COD, nitrogen content and PH value at different water depths, realizes instrument automation, avoids secondary pollution, saves costs, and improves the accuracy and comprehensiveness of water quality monitoring. , saving a lot of manpower, material and financial resources. At the same time, the instrument supports on-site measurement and remote monitoring, especially suitable for field water quality measurement; the instrument uses multiple channels to measure different water quality parameters, effectively reducing the interference of irrelevant factors on water quality measurement, making the measurement more Accurate and fast; the introduction of the SD card data dynamic access function makes the operation easier and more convenient. At the same time, the instrument is small in size and easy to carry, and the price is low, which is closer to the actual application, meets the needs of industry and individuals to a certain extent, and helps to further improve the environment The pollution problem has high social practical value and environmental protection value.

Claims (7)

1. The multifunctional underwater water quality automatic monitoring system based on the clustering algorithm is characterized in that: the monitoring system comprises an upper computer and a monitor, wherein the upper computer comprises a Labview interface design part and a Python data processing part, the upper computer designs a GUI interface through the Labview, calls a Python function by means of an internal interface function and adopts a clustering algorithm to complete sample data processing, and the monitor comprises a telescopic fixed support, a box body, a lower computer, a sliding module and a power supply module; wherein:

the lower end of the telescopic fixing bracket is fixed in water and used for fixing the whole monitor;

the box body is sealed, a hole is formed in the top end of the box body, a measuring switch is arranged outside the box body, a waterproof layer is arranged on the inner wall of the box body, and the lower computer and the power supply module are located in the waterproof layer;

the lower computer is arranged in the box body and is used for controlling the sliding module, collecting water quality parameter data and uploading the collected water quality parameter data to the upper computer;

the sliding module is used for connecting the box body and the telescopic fixing bracket;

the power module is arranged in the box body and comprises a direct current power supply driving module and a voltage amplifying module, the output end of the direct current power supply driving module is connected with the input end of the voltage amplifying module, and the output end of the voltage amplifying module is respectively connected with the power input end of each module to supply power for each module of the monitor;

the lower computer controls the sliding module to enable the box body to move up and down in the vertical direction, and then a plurality of water quality parameters of different water depths are collected;

the lower computer comprises a single chip microcomputer, a multi-channel spectrophotometric water quality measuring module and a data access module, wherein water quality information measured by the multi-channel spectrophotometric water quality measuring module is converted into an electric signal and then is output to the single chip microcomputer for processing, the data is output to the data access module after being processed by the single chip microcomputer, and the data is read from the data access module by the upper computer;

the single chip microcomputer comprises a control unit, an external RTC (real time clock), an external crystal oscillator, a signal conditioning amplifying circuit and a multi-channel ADC (analog to digital converter) conversion module, the multi-channel spectrophotometric water quality measuring module comprises a fixed base, a two-channel spectrophotometric device, a test tube fixing device, a silicon photocell sensor 1, a silicon photocell sensor 2, an ultraviolet light source 1, an ultraviolet light source 2, a titration tank and a water valve, the two-channel spectrophotometric device is fixed on the fixed base, a light path on-off switch is arranged on the inner wall of the test tube fixing device, a test tube for containing a to-be-tested agent is placed in the test tube fixing device, the opening of the test tube is right opposite to the top of a box body, the water valve is arranged at the upper end of the test tube fixing device, when the to-be-; the external RTC and the external crystal oscillator are connected with the control unit, the control unit controls the two ultraviolet light sources to emit ultraviolet light with different intensities and frequencies through outputting PWM (pulse width modulation) waves, the ultraviolet light irradiates to the corresponding silicon photocell sensors through internal channels of the spectrometer respectively, then optical signals are converted into electric signals through the silicon photocell sensors, the electric signals are amplified by the input signal conditioning and amplifying circuit, and finally the electric signals are input into the control unit through the multi-channel ADC conversion module for processing.

2. The multifunctional underwater water quality automatic monitoring system based on the clustering algorithm as claimed in claim 1, wherein: scalable fixed bolster includes two sections upper and lower stainless steel iron pipes, and the upper segment is the telescopic joint, and the lower section is the fixed knot, and the telescopic joint lower extreme has a spring bolt, and it has a plurality ofly and spring bolt complex fixed orifices to distribute on the fixed knot, and when inserting the spring bolt different fixed orifices, whole scalable fixed bolster has different heights, and telescopic joint and fixed knot pass through spring bolt and fixed orifices fixed connection.

3. The multifunctional underwater water quality automatic monitoring system based on the clustering algorithm as claimed in claim 1, wherein: the data access module comprises an SD card data dynamic access module and an EEPROM data cache module, the SD card data dynamic access module is respectively in bidirectional connection with the control unit and the upper computer, and the EEPROM data cache module is connected with the control unit.

4. The multifunctional underwater water quality automatic monitoring system based on the clustering algorithm as claimed in claim 1, wherein: the sliding module comprises a stepping motor, a fixed pulley and a movable pulley, the fixed pulley is fixed at the upper end of the telescopic fixing support, the movable pulley is fixed on the outer side wall of one side, close to the telescopic fixing support, of the box body and is connected with the telescopic fixing support, the stepping motor is fixed on the upper portion of the box body, a traction wire led out by the fixed pulley is sleeved on an output shaft of the stepping motor, the stepping motor is controlled by the single chip microcomputer to rotate clockwise or anticlockwise, the box body is driven to move up and down in the vertical direction, and the movable pulley moves up and.

5. The monitoring method of the multifunctional underwater water quality automatic monitoring system based on the clustering algorithm as claimed in any one of claims 1 to 4, characterized by comprising the following steps:

(1) debugging the monitor before starting to measure: adjusting the length of the telescopic fixed support to a height suitable for the current water level, inserting the telescopic fixed support underwater, and stably fixing the telescopic fixed support;

(2) judging whether the measurement switch is turned on, if so, inputting the current water depth and the height value of each descending unit of the box body to the control unit, and executing the step (3); if not, continuing to execute the step (2);

(3) before monitoring is started, judging whether the current lowered height is larger than or equal to an input water depth value or not, if the current lowered height is found to be larger than or equal to the input water depth value before monitoring is started, indicating that the input in the step (2) is wrong, and returning to the step (2) to input the current water depth and the height value of each time that the box body is lowered; if not, executing the step (4);

(4) the control unit controls the stepping motor to rotate forwards, so that the box body descends by a unit height value, and different water quality parameter values of the current water depth position are monitored;

(5) after one measurement is finished, the control unit recalculates and judges whether the current lowered height is larger than or equal to the input water depth value, if not, the instrument does not reach the water bottom, and the step (4) is returned to execute; if yes, executing the step (6);

(6) the control unit controls the stepping motor to rotate reversely, so that the box body rises by a unit height value;

(7) the control unit calculates and judges whether the current raised height is larger than or equal to the input water depth value, if not, the instrument does not reach the water surface, and the step (6) is returned to execute; if so, the box body is indicated to reach the water level, the control unit stores the measurement result into the data access module, and the measurement is finished;

(8) after the measurement is finished, the data access module is connected to an upper computer, the Labview reads data stored in the data access module through an internal function interface to serve as a sample data set to be processed, calls a Python function through an internal interface function to complete sample data processing through a clustering algorithm, and displays the sample data through a GUI interface designed by the Labview.

6. The method for monitoring the multifunctional underwater water quality automatic monitoring system based on the clustering algorithm as claimed in claim 5, wherein the method for monitoring the plurality of water quality parameter values of the current water depth position in the step (4) comprises the following steps:

(1) when the box body reaches the preset position, the control unit controls the water valve to be opened, so that external water quality flows into the test tube along the water inlet above the box body, and if the water level of the poured water quality reaches above the water valve, the control unit controls the water valve to be closed, so that the purpose of taking a water sample is achieved;

(2) the control unit controls the titration tank to titrate the current water sample by using different reagents according to the water quality parameters needing to be measured currently, and closes the titration tank after titration is finished;

(3) the control unit outputs PWM waves with frequency i to an ultraviolet light source to enable the ultraviolet light source to emit ultraviolet light with corresponding frequency and intensity, a spectrophotometry method is adopted to measure a current water quality parameter alpha, two silicon photocell sensors are used for measuring N times respectively, an average value of the two silicon photocell sensors is taken after the maximum value and the minimum value are removed, and current water quality parameter results are stored in an SD card respectively; wherein the water quality parameter alpha is COD concentration, nitrogen content or PH value of the water quality, and i represents the frequency of PWM wave required for obtaining the current water quality parameter alpha.

7. The monitoring method of the multifunctional underwater water quality automatic monitoring system based on the clustering algorithm according to claim 5, wherein the upper computer in the step (8) processes the sample data by adopting the clustering algorithm, and specifically comprises the following steps:

the system directly calls a Python interface function to open a Python console by means of Labview and opens the Python console by means of LabviewThe KMeans function coding clustering algorithm program in the Python built-in scimit-spare packet needs to be firstly imported into water quality monitoring data sets which are input to different positions and different water depths of an upper computer part through a data access module as a sample data set D after an interface is normally connected and a Python control console is opened, k samples are randomly selected from the D as an initial mean vector through a k value input by a user, the distance between each sample data in the data set and each sub-vector in the mean vector is calculated, and a distance measurement function adopts a Gaussian function to calculate: d_ij＝||x_j-u_i||₂I is more than or equal to 1 and less than or equal to k, j is more than or equal to 1 and less than or equal to m, wherein d_ijRepresenting the Gaussian distance, x, between the sample data and the mean vector_jRepresents the sample data to be divided, u_iRepresenting the cluster of samples formed by clustering, and sampling x_jDividing the data into clusters with the nearest distance, then recalculating a new mean vector, judging whether the mean vector is consistent with the previous mean vector or not, if so, finishing clustering, dividing the sample data set into k clusters, and if not, repeating the above calculation steps until the mean vector is not changed any more.

Images