River water quality monitoring system based on redundant sensor

Abstract

The invention discloses a river water quality monitoring system based on redundant sensors, and belongs to river water quality monitoring systems, the river water quality monitoring system based on redundant sensors is characterized in that when water quality parameters are measured, two data transmission stations are arranged at one water quality measuring point of a river, and for each water quality parameter, the data transmission stations are connected with the data transmission stations; three water quality sensors are used for measurement, two of the three water quality sensors are the same in manufacturer and model, and the two water quality sensors are connected with a data transmission station; the manufacturer and the model of the third water quality sensor are not limited, and the third water quality sensor is connected with the other data transmission station. The design of redundant sensors is adopted, when one water quality sensor is damaged during underwater work, other water quality sensors at the point position can continue to work, and meanwhile the redundant sensors can provide multiple pieces of required water quality basic parameter data for the credibility calculation unit of the data processing server.

Description

Technical Field

[0001] This invention relates to river water quality monitoring systems, and more particularly to river water quality monitoring systems based on redundant sensors. Background Technology

[0002] With the development of industry and agriculture, many regions have long experienced the direct or indirect inflow of industrial wastewater, domestic sewage, and agricultural drainage containing toxic and harmful substances into rivers. When these toxic and harmful substances accumulate in large quantities in a section of a river, exceeding the river's own purification capacity, it will cause changes in aquatic communities and water quality deterioration, endangering the production and living environment upon which humans depend. Water quality monitoring is a common way to observe whether a river is polluted. To this end, some water quality monitoring systems have been developed, such as those with announcement numbers CN112946227B and CN120121804A. However, these water quality monitoring methods have limitations in use. The water quality sensors are often single-type, and the probes of these sensors are easily damaged by impacts from hard objects such as rocks underwater, leading to a loss of monitoring function. Furthermore, river pollution exhibits a characteristic where the degree of pollution varies with runoff flow, making data measured by a single water quality sensor unreliable. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a river water quality monitoring system based on redundant sensors that can overcome or at least partially solve the above problems.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] A river water quality monitoring system based on redundant sensors includes: a water quality sensor, a data transmission station, a data processing server, and a central control cloud platform. The water quality sensor is wired to the data transmission station, the data transmission station is wirelessly connected to the data processing server, and the central control cloud platform is wirelessly connected to the data processing server.

[0006] The feature is that two data transmission stations are set up at a water quality measurement point in the river, and three water quality sensors are used to measure each water quality parameter. Two of the water quality sensors are manufactured by the same manufacturer and have the same model. The two water quality sensors are connected to a data transmission station. Because they have the same characteristics, the measured values ​​of the two water quality sensors are similar.

[0007] The third water quality sensor, whose manufacturer and model may differ from those of the two water quality sensors, is connected to another data transmission station. The measurement value of the third water quality sensor is used as a comparison parameter to compare with the measurement values ​​of the two water quality sensors from the same manufacturer, in order to determine the water quality parameters and their reliability at the water quality measurement point in the river.

[0008] Furthermore, the water quality sensor can be any one of five types of sensors: pH sensor, conductivity sensor, dissolved oxygen sensor, turbidity sensor, and temperature sensor. The water quality sensor can detect the water quality parameters at the water quality measurement point.

[0009] Furthermore, the data transmission station can be a transmitter or a wireless signal transceiver device for a water quality sensor.

[0010] The data transmission station can also be a circuit board that can be connected to peripheral circuits to realize data transmission and control functions;

[0011] When the data transmission station is a circuit board that can connect to peripheral circuits to realize data transmission and control functions, the circuit board needs to be installed in a waterproof protective box.

[0012] Furthermore, the data processing server can execute a reliability algorithm for water quality parameter measurements, an algorithm for triggering an alarm at a water quality measurement point in a given area due to exceeding water quality parameters, and an algorithm for triggering an alarm at multiple water quality measurement points due to exceeding the same type of water quality parameter. The server then wirelessly transmits the calculation results to the central control cloud platform.

[0013] Further, the method for a river water quality monitoring system based on redundant sensors according to claim 1 includes:

[0014] Step 1: Select water quality measurement points L at the upstream and downstream of major residential and industrial areas where sewage flows into rivers, at the inlets and outlets of lakes and reservoirs, and at the confluence of river tributaries and the main stream. Use three water quality sensors (a, b, and c) to measure the water quality parameter x, where x can be any one of the following parameters: pH, conductivity, dissolved oxygen, turbidity, and temperature. L-a x L-b and x L-c These are the measurement values ​​of sensors a, b, and c at water quality measurement point L for x, respectively.

[0015] Step 2: The data processing server executes a reliability algorithm for the water quality parameter measurements. This algorithm compares the measurements from the three water quality sensors to determine the water quality parameters and their reliability at the measurement point.

[0016] Step 3: The data processing server executes an algorithm for the water quality measurement points of a location's water quality to calculate whether a water pollution event has occurred in a small-range river section due to the exceeding of water quality parameters;

[0017] Step 4: The data processing server executes an algorithm for multiple water quality measurement points of the same water quality parameter to calculate whether a water pollution event has occurred in a large-range river section due to the exceeding of the same water quality parameter;

[0018] Step 5: On the total control cloud platform, display the river water quality parameters with credibility and the quality of the corresponding sensor working status through multiple function modules.

[0019] Furthermore, the algorithm in Step 2 is as follows:

[0020] When the measured value x of water quality parameter x L-a and x L-b satisfy |x L-a - x L-b | ≤ ε, it is said that x L-a is close to x L-b with ε, denoted as x L-a ≈ ε x L-b ; when the measured values x L-a and x L-b of water quality parameter x do not satisfy |x L-a - x L-b | ≤ ε, it is said that x L-a is far from x L-b by more than ε, denoted as

[0021] Among them, the value of ε is selected through experiments. The specific method is:

[0022] When the measured values of the two sensors measuring water quality parameter x at the water quality measurement point L of the river at times t1, t2, and t3 are x L-a (t1) and x L-b (t1), x L-a (t2) and x L-b (t2), x L-a (t3) and x L-b (t3),

[0023] ε1 = max{|x L-a (t1) - x L-b (t1)|, |x L-a (t2) - x L-b (t2)|, |x L-a (t3) - x L-b (t3)|}

[0024] When the measured values of the water quality parameter x by three sensors a, b, and c at the water quality measurement point L of the river at times t1, t2, and t3 are x L-a (t1), x L-b (t1), and x L-c (t1), x L-a (t2), x L-b (t2), and x L-c (t2), x L-a (t3), x L-b (t3), and x L-c (t3),

[0025] ε2 = max{|x L-a (t1) - x L-c (t1)|, |x L-a (t2) - x L-c (t2)|, |x L-a (t3) - x L-c (t3), |x L-b (t1) - x L-c (t1)|, |x L-b (t2) - x L-c (t2)|, |x L-b (t3) - x L-c (t3)|}

[0026] Three groups of experiments are conducted through two probes of the water quality sensor to obtain the ε value;

[0027] The specific method for selecting the ε value through experiments can be applied to monitor three or four or five water quality parameters using water quality sensors produced by the same manufacturer. Here, the water quality parameters are the acidity and alkalinity, conductivity, dissolved oxygen, turbidity, and temperature of the water quality;

[0028] When ε is the ε value at the water quality measurement point L, the credibility of the measured value of the water quality parameter x at this water quality measurement point can be defined in the following four ways:

[0029] Ⅰ. When x L-a ≈ ε1 x L-b , and at the same time x L-a ≈ ε2 x L-c and x L-b ≈ ε2 x L-c , then it is said that the measured value of the water quality parameter x at the water quality measurement point L is highly credible, denoted as C(x L ) = 3;

[0030] Ⅱ. When x L-a ≈ ε1 x L-b , and at the same time or The water quality measurement value of water quality parameter x at water quality measurement point L is said to be of moderate reliability, denoted as C(x). L ) = 2;

[0031] III. When There exists x L-a ≈ ε2 x L-c or x L-b ≈ ε2 x L-c If the measured value of water quality parameter x at water quality measurement point L is deemed to be of low reliability, denoted as C(x)... L ) = 1;

[0032] IV. When at the same time or Then the measured value of water quality parameter x at water quality measurement point L is said to be unreliable, denoted as C(x). L ) = 0;

[0033] At water quality measurement point L, the measured value of water quality parameter x is known. L-a x L-b and x L-c In this case, the following algorithm can be used to determine the reliability of the measurement values ​​in the data processing server:

[0034] a. When ε2≥ε1>0 is true, proceed to step b;

[0035] When ε2≥ε1>0 is not true, request the central control cloud platform to have the administrator input a reasonable allowable error value between the water quality sensor measurement values;

[0036] b. Initialize integer variables n, v1, v2, and v3 to 0;

[0037] c. When x L-a ≈ ε1 x L-b When it is established, v1 is assigned the value 1;

[0038] When x L-a ≈ ε1 x L-b If the condition is not met, v1 is assigned the value 0, and the process proceeds to step d.

[0039] d. When x L-a ≈ ε2 x L-c When it is established, v2 is assigned the value 1;

[0040] When x L-a ≈ ε2 x L-c If the condition is not met, v2 is assigned the value 0, and the process proceeds to step e.

[0041] e. When x L-b ≈ ε2 x L-c When it is set, v3 is assigned the value 1.

[0042] When x L-b ≈ ε2 x L-c If the condition is not met, v3 is assigned the value 0, and step f is executed.

[0043] f. Let n = v1*4 + v2*2 + v3;

[0044] g. When n=7, the measured value of water quality parameter x is highly reliable;

[0045] h. When n > 3, the measured value of water quality parameter x is of moderate reliability;

[0046] i. When n > 0, the measured value of water quality parameter x is of low confidence.

[0047] When n > 0 is not true, the measured value of water quality parameter x is unreliable;

[0048] At this point, at a water quality measurement point L in the river, there are three measured values ​​for a water quality parameter x. L-a x L-b and x L-c Based on the reliability C(x) of the measured value L The value of water quality parameter x varies depending on the difference in ) L Select according to the following rules:

[0049] When C(x) L When x ≥ 2, L =(x L-a+ x L-b ) / 2;

[0050] When C(x) L When x = 1, L =x L-c ;

[0051] When C(x) L When x = 0, L =0.

[0052] Furthermore, the algorithm in step three is as follows:

[0053] x L For the measured value of water quality parameter x at the river water quality measurement point L, x L (t i ) is x L In t i The measured value at time;

[0054] Event E is the event that the water quality of the river exceeds the standard. P(E = 1) represents the probability of the occurrence of the event that the water quality exceeds the standard, and P(E = 0) represents the probability of the event that the water quality does not exceed the standard. According to the national standard, some water quality parameters such as pH have two limit values, an upper limit and a lower limit, while some water quality parameters such as turbidity have only an upper limit value. Therefore, for pH, a measurement value that is too low or too high means that the water quality exceeds the standard;

[0055] P(C(x L ) = k) represents the probability that the water quality parameter x L is of k - level credibility, where k can take values 3, 2, 1, and 0, which respectively represent the probabilities that the water quality parameter x L is of high credibility, medium credibility, low credibility, and non - credibility. P(C(x L ) ≥ k) represents the probability that the water quality parameter x L is of no less than k - level credibility. When k takes the value of 2, P(C(x L ) ≥ 2) represents the probability that the water quality parameter x L is of no less than 2 - level credibility, and its calculation formula is:

[0056] P(C(x L ) ≥ 2) = P(C(x L ) = 3) + P(C(x L ) = 2)P(E = 1, C(x L ) ≥ k) represents the probability that the credibility of the measured value of the water quality parameter x L is not less than k and the event of water quality exceeding the standard occurs simultaneously;

[0057] In a set of measured values of the water quality parameter x L , the conditional probability of the occurrence of the event that the water quality exceeds the standard under the condition that the credibility of the measured value is not less than k is expressed as P(E = 1丨C(x L ) ≥ k). According to the calculation formula of conditional probability

[0058] P(E = 1丨C(x L ) ≥ k) = P(E = 1, C(x L ) ≥ k) / P(C(x L ) ≥ k)

[0059] The algorithm for judging the alarm caused by the exceeding of the water quality parameter x at the river water quality measurement point L can be determined by the following method. Its basic idea is that when the measured value of the water quality parameter with a certain credibility exceeds the limit value for a period of time, it is considered that a water pollution event caused by the exceeding of the water quality parameter x has occurred at the river water quality measurement point L;

[0060] Let x L (t i) The measured value of water quality parameter x at the current moment, m c The number of times the measured value of water quality parameter x is not lower than the k - level confidence, m Ec The number of times water quality parameter x exceeds the limit when the confidence of the measured value is not lower than k, I L The number of consecutive times water quality parameter x exceeds the limit when the confidence of the measured value is not lower than k, which is a constant input by the user from the total control cloud platform, δ L The conditional probability threshold of water quality exceeding the standard when the confidence of the measured value of water quality parameter x is not lower than k, which is also a constant input by the user from the total control cloud platform;

[0061] a. Assign i, m c , m Ec is 0;

[0062] b. i = i + 1;

[0063] c. When C(x L (t i )) ≥ k holds, m c = m c + 1, enter step d;

[0064] When C(x L (t i )) ≥ k does not hold, enter step e;

[0065] d. When x L (t i ) exceeds the limit, m Ec = m Ec + 1, enter step e;

[0066] When x L (t i ) does not exceed the limit, enter step e;

[0067] e. When the loop count i is greater than I L ), enter step f;

[0068] When the loop count i is not greater than I L ), enter step a;

[0069] f. Calculate P(E = 1丨C(x L ) ≥ k) = (m Ec / i) / (m c / i);

[0070] g. When P(E = 1丨C(x L ) ≥ k) ≥ δ L ), give an alarm on the total control cloud platform and enter step b;

[0071] When P(E = 1|C(x L ) ≥ k) < δ L then enter step a.

[0072] Furthermore, the algorithm in step four is as follows:

[0073] For the water quality parameter x, the measured values and their credibility of adjacent multiple water quality measurement points at different times are stored in a four-dimensional array G, and its elements are represented by [Lj][t i [x Lj (t i )][C(x Lj (t i ))]. Lj represents the jth water quality measurement point arranged in a certain section of the river, t i represents the time of measurement, x Lj (t i ) represents the measured value of the water quality parameter x at the jth water quality measurement point at time t i , and C(x Lj (t i )) represents the credibility of the measured value of the water quality parameter x at the jth water quality measurement point; by comparing whether the measured value x Lj (t i ) of different water quality measurement points Lj exceeds the national standard limit, determine whether a large-scale river section water pollution event occurs.

[0074] Furthermore, the data processing server sends the calculation results of steps two, three, and four to the total control cloud platform wirelessly, and deletes the data every quarter or half year.

[0075] Furthermore, the total control cloud platform has a system display screen, a display module, a control module, an alarm module, and an equipment maintenance module. The display module, the control module, the alarm module, and the equipment maintenance module are connected to the system display screen through wire harnesses and are interconnected to form the total control cloud platform;

[0076] The total control cloud platform can receive the water quality parameters and results sent by the storage edge server wirelessly, and can display the water quality parameters and their credibility of the corresponding water quality measurement points on the system display screen in the form of charts or numbers.

[0077] Furthermore, the display module has a graphical display module, a single-point query display module, and a multi-point query display module. The single-point query display module and the multi-point query display module can convert the change information of the water quality parameters into a graphical curve and display it on the system display screen.

[0078] Preferably, the graphical curve has lines of four colors: green, blue, yellow, and red. Green lines represent water quality parameters with high confidence, blue lines represent water quality parameters with medium confidence, yellow lines represent water quality parameters with low confidence, and red lines represent water quality parameters with no confidence.

[0079] Furthermore, the control module can issue commands to retrieve water quality parameters stored in the data processing server and to set commands for transmitters or smart water quality sensors. When the water quality sensor is a non-smart sensor, it can only receive setting commands from the transmitter. In this case, the central control cloud platform can only set the transmitter and not the water quality sensor.

[0080] Furthermore, the equipment maintenance module can display the data processing server, data transmission station, and water quality sensor and their working status on the system display screen of the central control cloud platform.

[0081] Preferably, the system display screen allows selection of a water quality parameter. Clicking the selection box for the corresponding confidence level will display all sensors whose measured values ​​fall within the selected confidence level.

[0082] When the reliability of the water quality measurement value at the water quality measurement point is 0, it indicates that all three sensors have large errors and should be maintained.

[0083] When the reliability of the water quality measurement value at the water quality measurement point is 1, it indicates that two of the three sensors of the same manufacturer and model have a large error and should be maintained.

[0084] When the reliability of the water quality measurement value at the water quality measurement point is 2, it indicates that the third sensor among the three sensors has a large error, and maintenance personnel can perform maintenance when passing by.

[0085] When the reliability of the water quality measurement value at the water quality measurement point is 3, it means that all three sensors at the water quality measurement point are in good working condition and no maintenance is required.

[0086] Furthermore, the river water quality monitoring system based on redundant sensors can be used for water quality monitoring of water supply networks and drainage networks.

[0087] By adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0088] 1. This invention measures a water quality parameter using three water quality sensors. If one water quality sensor is damaged, the remaining water quality sensors can still work normally, allowing monitoring to continue.

[0089] Meanwhile, for a single water quality parameter, there are three measurement results. The credibility calculation unit of the data processing server can perform credibility calculations on the three measurement results, calculate the value of the water quality parameter and its credibility, and improve the accuracy of the measurement results.

[0090] 2. By using the water quality exceedance calculation unit in the data processing server, the measured water quality parameters can be calculated to determine if water quality exceeds the standard. If a water pollution event occurs, an alarm will be issued to remind regulatory personnel to pay attention to the abnormal water quality.

[0091] 3. By setting up the central control cloud platform, regulatory personnel can intuitively understand the water quality parameters of each water quality measurement point and the working status of the corresponding sensors, enabling them to identify problems and take countermeasures immediately. They can also remotely set parameters for water quality sensors or data transmission stations, facilitating coordinated management.

[0092] In summary, this invention, through redundant sensor design, can avoid measurement interruption caused by the failure of one water quality sensor; at the same time, it can also provide multiple measurement data to the reliability calculation unit, making the measurement results more accurate; the design of the water quality exceedance calculation unit in the data processing server can alert regulatory personnel to abnormal water quality conditions; and through the setting of the central control cloud platform, the measurement results can be displayed intuitively, facilitating coordination and management by regulatory personnel. Attached Figure Description

[0093] Figure 1 This is a schematic diagram of the process in the river water quality monitoring system based on redundant sensors proposed in this invention, when the number of each type of water quality sensor is three.

[0094] Figure 2 This is a logic diagram of the data processing server reliability calculation unit in the river water quality monitoring system based on redundant sensors proposed in this invention.

[0095] Figure 3 This is a logic diagram of the water quality exceedance calculation unit of the data processing server in the river water quality monitoring system based on redundant sensors proposed in this invention.

[0096] Figure 4 This is a schematic diagram showing the array G on the central control cloud platform of the river water quality monitoring system based on redundant sensors proposed in this invention;

[0097] Figure 5 This is a flowchart illustrating the process of a river water quality monitoring system based on redundant sensors proposed in this invention, when the number of water quality sensors is N. Detailed Implementation

[0098] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0099] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0100] Example 1:

[0101] Reference Figure 1 , Figure 2 and Figure 5 A river water quality monitoring system based on redundant sensors includes: water quality sensors, data transmission stations, data processing servers, and a central control cloud platform. The water quality sensors are wired to the data transmission stations, the data transmission stations are wirelessly connected to the data processing servers, and the central control cloud platform is wirelessly connected to the data processing servers.

[0102] Water quality sensors are used to detect changes in the water quality parameters of a river and send the measurement results to a data transmission station. Signal transmission between the water quality sensor and the data transmission station is wired. Water quality sensors include, but are not limited to, five conventional parameter sensors—pH, conductivity, dissolved oxygen, turbidity, and temperature sensors—and are also suitable for composite sensors integrating two or more sensing units on a single sensor. For example, intelligent pH sensors from multiple manufacturers can simultaneously measure both pH and temperature, two water quality parameters.

[0103] This method measures a water quality parameter at a single point in a river using three medium- to low-cost sensors. Two of these sensors are from the same manufacturer and are of the same model, possessing similar measurement characteristics. Furthermore, the output of each water quality sensor is connected to only one data transmission station.

[0104] The data transmission station receives and processes the output signals from the water quality sensor and wirelessly transmits water quality parameter values ​​to the data processing server. Simultaneously, the data processing server can also send parameter settings to the intelligent water quality sensor via the data transmission station. The data transmission station can be a transmitter or wireless transceiver device associated with the water quality sensor, or it can be a circuit board that can connect to external circuits to achieve data transmission and control functions. When the data transmission station is a circuit board that can connect to external circuits to achieve data transmission and control functions, the circuit board must be installed in a waterproof protective box.

[0105] This method allows setting up two data transmission stations at a single water quality measurement point in a river. For each measurement parameter, one data transmission station connects to two water quality sensors of the same manufacturer and model, while the other connects to a third water quality sensor. There are no restrictions on the manufacturer or model of the third water quality sensor. If needed, each data transmission station can connect to more water quality sensors, provided the hardware interface requirements are met, but the connection rules described above must be followed. For example, when connecting two different water quality sensors to a data transmission station, water quality sensors 2a, 2b, 3a, and 3b can be connected to one data transmission station 2a, while the remaining sensors 2c and 3c can be connected to another data transmission station 2b; or, water quality sensors 4a, 4b, and 5c can be connected to data transmission station 3a, while water quality sensors 5a, 5b, and 4c can be connected to data transmission station 3b.

[0106] The data processing server receives water quality parameters from the data transmission station wirelessly. It then executes algorithms to determine the reliability of the measured water quality parameters, to determine if an alarm is triggered at a single water quality measurement point due to excessive water quality parameters, and to determine if alarms are triggered at multiple measurement points due to excessive levels of the same water quality parameter. The results are then wirelessly transmitted to the central control cloud platform. Due to limited storage capacity, the data processing server automatically clears algorithms executed quarterly or semi-annually.

[0107] The central control cloud platform receives water quality parameters and calculation results from the data processing server wirelessly and displays these parameters in charts and other formats using multiple functional modules. Furthermore, the platform issues commands to retrieve water quality parameters stored on the data processing server and to configure transmitters or smart water quality sensors. It's important to note that when the water quality sensor is not smart, it can only receive configuration commands from the transmitter; in this case, the central control cloud platform can only configure the transmitter, not the sensor itself.

[0108] In river water quality monitoring, monitoring points are typically selected upstream and downstream of major residential and industrial areas where sewage flows into the river, at the inlets and outlets of lakes and reservoirs, and at the confluence of tributaries and the main stream. When measuring a specific water quality parameter at a monitoring point, three sensor probes need to be placed in the water. Two probes of the same manufacturer and model are placed close together, close enough not to interfere with their individual measurements. The third sensor probe is placed at a certain distance in the direction of river flow. When all three sensors are functioning normally, the measurements from the two sensors of the same manufacturer and model will be very close due to their similar characteristics, but also not significantly different from the measurement from the third sensor. The third sensor can also be of the same manufacturer and model as the side-by-side sensors, but due to the slightly different measurement location, its measurement will differ more from the first two. Placing the third sensor at a certain distance in the direction of river flow prevents all three sensors from being damaged simultaneously by impact.

[0109] To monitor water quality in more locations within a given measurement point, N water quality sensors (N being an integer greater than or equal to 3) can be used to measure a single water quality parameter. Two of these sensors must be from the same manufacturer and of the same model, possessing similar measurement characteristics; the remaining sensors have no manufacturer or model restrictions. B data transmission stations and A data processing servers are configured (A and B are both integers greater than or equal to 2). Figure 5 As shown;

[0110] It should be noted that although this invention provides an economical and practical method for low- to medium-cost water quality sensor applications while still maintaining high accuracy in measurements, the method is also applicable to high-precision sensors. In certain river sections, due to the complex underwater environment prone to rapids and eddies, even high-precision sensors sometimes struggle to function ideally or are easily damaged. It is reasonable to believe that applying the measurement method based on redundant sensors described herein can also improve the accuracy of the monitored water quality parameters. Without making significant improvements to the method in this invention, the aforementioned areas should be considered within the scope of protection of this patent.

[0111] Let L be the water quality measurement point on a river for water quality parameter x, where x is any parameter such as pH, conductivity, dissolved oxygen, turbidity, and temperature. Then x L-a x L-b and x L-c These are the measurement values ​​of sensors a, b, and c at water quality measurement point L for x.

[0112] When the measured value of water quality parameter x is x L-a and x L-b Satisfy | x L-a -x L-b When |≤ε, x is calledL-a Let ε be close to x L-b , denoted as x L-a ≈ ε x L-b , when the measured value x of the water quality parameter x L-a and x L-b do not satisfy |x L-a -x L-b | ≤ ε, x is said to be L-a far from x L-b above ε, denoted as

[0113] where the value of ε is selected through experiments, and the specific method is as follows:

[0114] When the measured values of the water quality parameter x by two sensors a and b at the water quality measurement point L of the river at times t1, t2, and t3 are x L-a (t1) and x L-b (t1), x L-a (t2) and x L-b (t2), x L-a (t3) and x L-b (t3),

[0115] ε1 = max{|x L-a (t1)-x L-b (t1)|,|x L-a (t2)-x[[ID=5o]] L-b (t2)|,|x L-a (t3)-x L-b (t3)|}

[0116] When the measured values of the water quality parameter x by three sensors a, b, and c at the water quality measurement point L of the river at times t1, t2, and t3 are x L-a (t1), x L-b (t1) and x L-c (t1), x L-a (t2), x L-b (t2) and x L-c (t2), x L-a (t3), x L-b (t3) and x L-c (t3),

[0117] ε2 = max{|x L-a (t1)-x L-c (t1)|,|x L-a (t2)-x L-c (t2)|,|x L-a (t3)-x L-c (t3),|x L-b(t1)-x L-c (t1)丨,丨x L-b (t2)-x L-c (t2)丨,丨x L-b (t3)-x L-c (t3)丨}

[0118] Three groups of experiments are conducted through two probes of the water quality sensor to obtain the ε value;

[0119] The specific method for selecting the ε value through experiments can be applied to monitor three or four or five water quality parameters using water quality sensors produced by the same manufacturer. Here, the water quality parameters are the acidity and alkalinity, conductivity, dissolved oxygen, turbidity, and temperature of the water;

[0120] Assume that the measured values of the two acidity and alkalinity sensor probes a and b at the water quality measurement point L of a certain river at times t1, t2, and t3 are 7.34 and 7.38, 7.32 and 7.36, 7.33 and 7.39 respectively. Then, according to the above formula, ε is 0.06. If at time t1, the measured values of the acidity and alkalinity sensor probes a and b are 7.33 and 7.38 respectively, since 0.05 < 0.06, it indicates that the measured value of the acidity and alkalinity sensor probe a is close to the measured value of the acidity and alkalinity sensor probe b with 0.06.

[0121] When ε is the ε value at the water quality measurement point L, the credibility of the measured value of the water quality parameter x at this water quality measurement point can be defined in the following four ways:

[0122] Ⅰ. When x L-a ≈ ε1 x L-b ,and at the same time x L-a ≈ ε2 x L-c and x L-b ≈ ε2 x L-c ,then it is said that the measured value of the water quality parameter x at the water quality measurement point L is highly credible, denoted as C(x L ) = 3;

[0123] Ⅱ. When x L-a ≈ ε1 x L-b ,and at the same time or ,then it is said that the measured value of the water quality parameter x at the water quality measurement point L is moderately credible, denoted as C(x L ) = 2;

[0124] Ⅲ. When there exists x L-a ≈ ε2 x L-c or x L-b ≈ ε2x L-c If the measured value of water quality parameter x at water quality measurement point L is deemed to be of low reliability, denoted as C(x)... L ) = 1;

[0125] IV. When at the same time or Then the measured value of water quality parameter x at water quality measurement point L is said to be unreliable, denoted as C(x). L ) = 0;

[0126] At water quality measurement point L, the measured value of water quality parameter x is known. L-a x L-b and x L-c In this case, the following algorithm can be used to determine the reliability of the measurement values ​​in the data processing server:

[0127] a. When ε2≥ε1>0 is true, proceed to step b;

[0128] When ε2≥ε1>0 is not true, request the central control cloud platform to have the administrator input a reasonable allowable error value between the water quality sensor measurement values;

[0129] b. Initialize integer variables n, v1, v2, and v3 to 0;

[0130] c. When x L-a ≈ ε1 x L-b When it is established, v1 is assigned the value 1;

[0131] When x L-a ≈ ε1 x L-b If the condition is not met, v1 is assigned the value 0, and the process proceeds to step d.

[0132] d. When x L-a ≈ ε2 x L-c When it is established, v2 is assigned the value 1;

[0133] When x L-a ≈ ε2 x L-c If the condition is not met, v2 is assigned the value 0, and the process proceeds to step e.

[0134] e. When x L-b ≈ ε2 x L-c When it is set, v3 is assigned the value 1.

[0135] When x L-b ≈ ε2 x L-c If the condition is not met, v3 is assigned the value 0, and step f is executed.

[0136] f. Let n = v1*4 + v2*2 + v3;

[0137] g. When n=7, the measured value of water quality parameter x is highly reliable;

[0138] h. When n > 3, the measured value of water quality parameter x is of moderate reliability;

[0139] i. When n > 0, the measured value of water quality parameter x is of low confidence.

[0140] When n > 0 is not true, the measured value of water quality parameter x is unreliable;

[0141] The algorithm described above is only one implementation for determining the reliability of water quality parameter x measurements. Other implementation algorithms exist. Any method that uses the concepts of this invention to calculate the reliability of measurements falls within the scope of protection of this invention.

[0142] At this point, at a water quality measurement point L in the river, there are three measured values ​​for a water quality parameter x. L-a x L-b and x L-c Based on the reliability C(x) of the measured value L The value of water quality parameter x varies depending on the difference in ) L Select according to the following rules:

[0143] When C(x) L When x ≥ 2, L =(x L-a+ x L-b ) / 2;

[0144] When C(x) L When x = 1, L =x L-c ;

[0145] When C(x) L When x = 0, L =0.

[0146] The calculation of water quality parameters and their reliability at a specific water quality measurement point in the river is completed on the data processing server, and the calculated results are wirelessly transmitted to the central control cloud platform. Only when the central control cloud platform issues a specified read command will the data processing server send the raw measurement values ​​from the three sensors corresponding to the water quality parameters measured at that point to the central control cloud platform; otherwise, the raw measurement values ​​are simply stored on the data processing server.

[0147] On the central control cloud platform, the graphical display module for water quality parameters is responsible for plotting the changing patterns of water quality parameters and displaying them graphically on the system screen. The single-point query display module displays the water quality parameter change curves over a specified time period based on the user-input water quality measurement points and parameters. Combined with redundant water quality sensor settings, the graphical display module also has the function of selectively displaying the reliability of water quality parameters at different times within a specified time period and statistical information on different reliability states.

[0148] The multi-point query display module can display water quality parameter change curves for multiple selected water quality measurement points at a specified time, based on the measurement parameters and measurement time input by the user. Combined with redundant water quality sensor settings, the module also has the function of selectively displaying the reliability of water quality parameters at multiple selected water quality measurement points at a specified time and statistical information on different reliability states.

[0149] On the water quality parameter curve, four colors—green, blue, yellow, and red—represent the four states of water quality parameters at selected measurement points at a specified time: high confidence, moderate confidence, low confidence, and unreliable. Since unreliable water quality parameters have a value of 0, they will be represented by a red dot or line segment on the horizontal axis of the curve. When the statistics information selection box is clicked, the system display will show the number of each of the four states—high confidence, moderate confidence, low confidence, and unreliable—and their respective proportions.

[0150] Example 2:

[0151] Reference Figure 3 , Figure 4 Based on Example 1, the difference lies in the following: On the central control cloud platform, the water quality parameter alarm module is used to alert river management personnel to abnormal water quality conditions when water quality parameters exceed standards. An alarm is issued when the measured value of a certain water quality parameter exceeds the upper limit or falls below the lower limit stipulated by the national standard. Especially when the alarm persists for a period of time or spreads along a certain range along the river, a water pollution incident can be considered to have occurred. The water quality alarms of this invention include two types: one is an alarm triggered by a single water quality measurement point exceeding a certain water quality parameter, indicating a water pollution incident at a river location or a small section of the river; the other is an alarm triggered by multiple water quality measurement points exceeding the same water quality parameter, indicating a water pollution incident over a larger section of the river.

[0152] To illustrate the algorithm for triggering an alarm at a single water quality measurement point due to an exceedance of a certain water quality parameter, the following notation is first given:

[0153] x L For the measured value of water quality parameter x at the river water quality measurement point L, x L (t i ) is xL The measured value at time t i ;

[0154] Event E is the event that the water quality of the river exceeds the standard. P(E = 1) represents the probability of the occurrence of the event that the water quality exceeds the standard, and P(E = 0) represents the probability of the event that the water quality does not exceed the standard. According to the national standard, some water quality parameters, such as acidity and alkalinity, have two limit values, an upper limit and a lower limit, while some water quality parameters, such as turbidity, have only an upper limit value. Therefore, for acidity and alkalinity, too low or too high measured values both mean that the water quality exceeds the standard;

[0155] P(C(x L ) = k) represents the probability that the water quality parameter x L is of k - level credibility, where k can take values 3, 2, 1, and 0, which respectively represent the probabilities that the water quality parameter x L is highly credible, moderately credible, lowly credible, and non - credible. P(C(x L )≥k) represents the probability that the water quality parameter x L is of no less than k - level credibility. If k takes the value of 2, P(C(x L )≥2) represents the probability that the water quality parameter x L is of no less than 2 - level credibility, and its calculation formula is:

[0156] P(C(x L )≥2) = P(C(x L ) = 3)+P(C(x L ) = 2)P(E = 1,C(x L )≥k) represents the probability that the measured value of the water quality parameter x L and the occurrence of the event that the water quality exceeds the standard occur simultaneously with the credibility of the measured value not less than k;

[0157] Among a set of measured values of the water quality parameter x L , the conditional probability of the occurrence of the event that the water quality exceeds the standard under the condition that the credibility of the measured value is not less than k is expressed as P(E = 1丨C(x L )≥k). According to the calculation formula of conditional probability

[0158] P(E = 1丨C(x L )≥k) = P(E = 1,C(x L )≥k) / P(C(x L )≥k)

[0159] The algorithm for judging the alarm caused by the exceeding of the water quality parameter x at the river water quality measurement point L can be determined by the following method. Its basic idea is that when the measured value of the water quality parameter with a certain credibility exceeds the limit value for a period of time, it is considered that a water pollution event caused by the exceeding of the water quality parameter x has occurred at the water quality measurement point L;

[0160] Let xL (t i ) is the measured value of water quality parameter x at the current moment, m c is the number of times that the measured value of water quality parameter x is not lower than the k - equal credibility, m Ec is the number of times that water quality parameter x exceeds the limit value when the credibility of the measured value is not lower than k, I L is the number of times that water quality parameter x continuously exceeds the limit value when the credibility of the measured value is not lower than k, which is a constant input by the user from the total control cloud platform, δ L is the conditional probability threshold of the water quality exceeding the standard event when the credibility of the measured value of water quality parameter x is not lower than k, and it is also a constant input by the user from the total control cloud platform;

[0161] a. Assign i, m c , m Ec to 0;

[0162] b. i = i + 1;

[0163] c. When C(x L (t i )) ≥ k holds, m c = m c + 1, and enter step d;

[0164] When C(x L (t i )) ≥ k does not hold, enter step e;

[0165] d. When x L (t i ) exceeds the limit value, m Ec = m Ec + 1, and enter step e;

[0166] When x L (t i ) does not exceed the limit value, enter step e;

[0167] e. When the loop count i is greater than I L , enter step f;

[0168] When the loop count i is not greater than I L , enter step a;

[0169] f. Calculate P(E = 1丨C(x L ) ≥ k) = (m Ec / i) / (m c / i);

[0170] g. When P(E = 1丨C(x L ) ≥ k) ≥ δ LWhen an alarm is given on the master control cloud platform, proceed to step b;

[0171] When P(E = 1|C(x L ) ≥ k) < δ L Proceed to step a.

[0172] The above algorithm can be further extended to achieve multiple purposes. For example, to calculate the conditional probability P(E = 1|C(x L ) = k) of a water quality exceeding the standard event when the confidence level of the measured value of water quality parameter x is k, just change P(E = 1|C(x L ) ≥ k) in the algorithm to P(E = 1|C(x L ) = k). According to the calculation formula of conditional probability, L P(E = 1|C(x

[0173] ) = k) = P(E = 1, C(x L ) = k) / P(C(x L ) = k) L )

[0174] The above algorithm combines the situations where the measured values of water quality parameters continuously exceed the limit and exceed the limit multiple times within a period of time, thus being able to reduce the influence of external interference on the measured values of water quality sensors. At the same time, the above algorithm is simple to implement and is more suitable for implementation on edge servers with relatively limited computing and storage resources.

[0175] The method for multiple water quality measurement points to give an alarm due to the same water quality parameter exceeding the standard is described below. For water quality parameter x, the measured values and their confidence levels of adjacent multiple water quality measurement points at different times are stored in a four-dimensional array G, and its elements are represented by [Lj][t i [x Lj (t i )][C(x Lj (t i ))]. Lj represents the j-th water quality measurement point arranged in a certain section of the river, t i represents the measurement time, x Lj (t i ) represents the measured value of water quality parameter x at the j-th water quality measurement point at time t i , and C(x Lj (t i )) represents the confidence level of the measured value of water quality parameter x at the j-th water quality measurement point.

[0176] On the alarm display page of the master control cloud platform, the array G is represented by the following two-dimensional graph, where the horizontal axis is Lj and t i , and the vertical axis is x Lj (t i ) and C(x Lj(t i Among them, Lj and t i On the same axis, time t at each Lj point is repeated multiple times, while the vertical axis uses short solid lines to represent the measured values ​​of water quality parameters, and the color of the solid lines indicates the reliability of the measured values. Similarly, green, blue, yellow, and red are used to represent four states of water quality parameters: high reliability, medium reliability, low reliability, and unreliable reliability.

[0177] Figure 4 The diagram shown is an example of an array G display graph, with four water quality measurement points L1, L2, L3, and L4. Each water quality measurement point underwent five measurements at times t1, t2, t3, t4, and t5. The dashed line in the graph represents the upper limit value x0 of the water quality parameter as specified by the national standard.

[0178] Figure 4 The measured value of water quality parameter x at water quality monitoring point L2 showed a significant increase compared to water quality monitoring point L1, exceeding the upper limit. Over time, the value initially increased and then decreased, indicating that sewage injection or the dumping of toxic substances into the river near the monitoring point caused the water quality to exceed the standard. Simultaneously, water quality exceeding the standard was also observed at water quality monitoring points L3 and L4, and the initial increase followed by a decrease in the measured value of water quality parameter x showed a certain lag, indicating that water quality monitoring points L3 and L4, located downstream of water quality monitoring point L1, were also affected by pollution near L1. Based on these observations, it can be inferred that a large-scale water pollution event occurred in the monitored river section from water quality monitoring points L2 to L4.

[0179] Example 3:

[0180] On the central control cloud platform, the equipment maintenance module will display all equipment and their status on a separate page. Selecting a water quality parameter and clicking the selection box for a confidence level will display all sensors whose measurements fall within the selected confidence level. When the confidence level of a water quality measurement at a certain point is 0, all three sensors have significant errors and should be maintained first. When the confidence level of a water quality measurement at a certain point is 1, two of the three sensors of the same manufacturer and model have significant errors and should also be maintained. When the confidence level of a water quality measurement at a certain point is 2, the third sensor has a significant error, and maintenance personnel can perform maintenance when passing by. When the confidence level of a water quality measurement at a certain point is 3, it indicates that all three sensors at that water quality measurement point are in good working order and require no maintenance.

[0181] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A river water quality monitoring system based on redundant sensors, including: The system includes a water quality sensor, a data transmission station, a data processing server, and a central control cloud platform. The water quality sensor is wired to the data transmission station, the data transmission station is wirelessly connected to the data processing server, and the central control cloud platform is wirelessly connected to the data processing server. The feature is that two data transmission stations are set up at a water quality measurement point in the river, and three water quality sensors are used to measure each water quality parameter. Two of the water quality sensors are manufactured by the same manufacturer and have the same model. The two water quality sensors are connected to a data transmission station. Because they have the same characteristics, the measured values ​​of the two water quality sensors are similar. The third water quality sensor, whose manufacturer and model may differ from those of the two water quality sensors, is connected to another data transmission station. The measurement value of the third water quality sensor is used as a comparison parameter to compare with the measurement values ​​of the two water quality sensors from the same manufacturer, in order to determine the water quality parameters and their reliability at the water quality measurement point in the river.

2. The river water quality monitoring system based on redundant sensors according to claim 1, characterized in that, The water quality sensor can be any one of five types of sensors: pH sensor, conductivity sensor, dissolved oxygen sensor, turbidity sensor, and temperature sensor. The water quality sensor can detect the water quality parameters at the water quality measurement point.

3. The river water quality monitoring system based on redundant sensors according to claim 1, characterized in that, The data transmission station can be a transmitter or a wireless signal transceiver device that is paired with the water quality sensor. The data transmission station can be a circuit board that can be connected to peripheral circuits to realize data transmission and control functions; When the data transmission station is a circuit board that can connect to peripheral circuits to realize data transmission and control functions, the circuit board needs to be installed in a waterproof protective box.

4. The river water quality monitoring system based on redundant sensors according to claim 1, characterized in that, The data processing server can execute a reliability algorithm for water quality parameter measurements, an algorithm for triggering an alarm at a water quality measurement point in a given area due to exceeding water quality parameters, and an algorithm for triggering an alarm at multiple water quality measurement points due to exceeding the same type of water quality parameter. The server then wirelessly transmits the calculation results to the central control cloud platform.

5. The method for a river water quality monitoring system based on redundant sensors according to claim 1, comprising: Step 1: Select water quality measurement points L at the upstream and downstream of major residential and industrial areas where sewage flows into rivers, at the inlets and outlets of lakes and reservoirs, and at the confluence of river tributaries and the main stream. Use three water quality sensors (a, b, and c) to measure the water quality parameter x, where x can be any one of the following parameters: pH, conductivity, dissolved oxygen, turbidity, and temperature. L-a x L-b and x L-c These are the measurement values ​​of sensors a, b, and c at water quality measurement point L for x, respectively. Step 2: The data processing server executes a reliability algorithm for the water quality parameter measurements. This algorithm compares the measurements from the three water quality sensors to determine the water quality parameters and their reliability at the measurement point. Step 3: The data processing server executes an algorithm that triggers an alarm at a water quality measurement point in a given location due to excessive water quality parameters, in order to calculate whether a water pollution event has occurred in a small section of the river. Step 4: The data processing server executes an algorithm that triggers alarms at multiple water quality measurement points due to the same water quality parameter exceeding the standard, in order to calculate whether a water pollution event has occurred in a large-scale river section; Step 5: The central control cloud platform displays reliable river water quality parameters and the corresponding sensor operating status through multiple functional modules.

6. The river water quality monitoring system based on redundant sensors according to claim 5, characterized in that, The algorithm in step two is as follows: When the measured value of water quality parameter x is x L-a and x L-b Satisfy | x L-a -x L-b When |≤ε, x is called L-a With ε close to x L-b , denoted as x L-a ≈ ε x L-b When the measured value of water quality parameter x is x L-a and x L-b Not satisfied | x L-a -x L-b When |≤ε, x is called L-a Stay away from x L-b Above ε, it is denoted as The value of ε was selected experimentally, specifically using the following method: When sensors a and b, which measure water quality parameter x, are at water quality measurement point L in the river at times t1, t2, and t3, the measured values ​​are x respectively. L-a (t1) and x L-b (t1), x L-a (t2) and x L-b (t2), x L-a (t3) and x L-b At (t3), ε1 = max{|x L-a (t1) - x L-b (t1)|, |x L-a (t2) - x L-b (t2)|, |x L-a (t3) - x L-b (t3)|} When sensors a, b, and c, which measure water quality parameter x, are at water quality measurement point L in the river at times t1, t2, and t3, the measured values ​​are x respectively. L-a (t1), x L-b (t1) and x L-c (t1), x L-a (t2), x L-b (t2) and x L-c (t2), x L-a (t3), x L-b (t3) and x L-c At (t3), ε2 = max{|x L-a (t1) - x L-c (t1)|, |x L-a (t2) - x L-c (t2)|, |x L-a (t3) - x L-c (t3), |x L-b (t1) - x L-c (t1)|, |x L-b (t2) - x L-c (t2)|, |x L-b (t3) - x L-c (t3)|} The ε value was obtained by conducting three sets of experiments using two probes of a water quality sensor. The ε value, determined through experiments, can be applied to monitor three, four, or five water quality parameters using water quality sensors manufactured by the same company. These water quality parameters are pH, conductivity, dissolved oxygen, turbidity, and temperature. When ε is the value of ε at water quality measurement point L, the reliability of the measured value of water quality parameter x at that water quality measurement point can be defined in the following four ways: Ⅰ. When x L-a ≈ ε1 x L-b Meanwhile, x L-a ≈ ε2 x L-c And x L-b ≈ ε2 x L-c If the measured value of water quality parameter x at water quality measurement point L is considered highly reliable, it is denoted as C(x). L ) = 3; II. When x L-a ≈ ε1 x L-b ,at the same time or The water quality measurement value of water quality parameter x at water quality measurement point L is said to be of moderate reliability, denoted as C(x). L ) = 2; III. When There exists x L-a ≈ ε2 x L-c or x L-b ≈ ε2 x L-c If the measured value of water quality parameter x at water quality measurement point L is deemed to be of low reliability, denoted as C(x)... L ) = 1; IV. When at the same time or Then the measured value of water quality parameter x at water quality measurement point L is said to be unreliable, denoted as C(x). L ) = 0; At water quality measurement point L, the measured value of water quality parameter x is known. L-a x L-b and x L-c In this case, the following algorithm can be used to determine the reliability of the measurement values ​​in the data processing server: a. When ε2≥ε1>0 is true, proceed to step b; When ε2≥ε1>0 is not true, request the central control cloud platform to have the administrator input a reasonable allowable error value between the water quality sensor measurement values; b. Initialize integer variables n, v1, v2, and v3 to 0; c. When x L-a ≈ ε1 x L-b When it is established, v1 is assigned the value 1; When x L-a ≈ ε1 x L-b If the condition is not met, v1 is assigned the value 0, and the process proceeds to step d. d. When x L-a ≈ ε2 x L-c When it is established, v2 is assigned the value 1; When x L-a ≈ ε2 x L-c If the condition is not met, v2 is assigned the value 0, and the process proceeds to step e. e. When x L-b ≈ ε2 x L-c When it is set, v3 is assigned the value 1. When x L-b ≈ ε2 x L-c If the condition is not met, v3 is assigned the value 0, and step f is executed. f. Let n = v1*4 + v2*2 + v3; g. When n=7, the measured value of water quality parameter x is highly reliable; h. When n > 3, the measured value of water quality parameter x is of moderate reliability; i. When n > 0, the measured value of water quality parameter x is of low confidence. When n > 0 is not true, the measured value of water quality parameter x is unreliable; At this point, at a water quality measurement point L in the river, there are three measured values ​​for a water quality parameter x. L-a x L-b and x L-c Based on the reliability C(x) of the measured value L The value of water quality parameter x varies depending on the difference in ) L Select according to the following rules: When C(x) L When x ≥ 2, L =(x L-a+ x L-b ) / 2; When C(x) L When x = 1, L =x L-c ; When C(x) L When x = 0, L =0.

7. The river water quality monitoring system based on redundant sensors according to claim 5, characterized in that, The algorithm in step three is as follows: x L For the measured value of water quality parameter x at the river water quality measurement point L, x L (t i ) is x L In t i The measured value at time; Event E represents the occurrence of water quality exceeding the standard in the river. P(E=1) represents the probability of the water quality exceeding the standard event occurring, and P(E=0) represents the probability of the water quality not exceeding the standard event occurring. According to national standards, some water quality parameters, such as pH, have two limits, an upper limit and a lower limit, while some water quality parameters, such as turbidity, only have one upper limit. Therefore, for pH, a measurement value that is too low or too high means that the water quality exceeds the standard. P(C(x L )=k) represents the water quality parameter x L It represents the probability of a confidence level of k, where k can take the values ​​3, 2, 1, and 0, which represent the water quality parameter x, respectively. L It represents the probabilities of high credibility, medium credibility, low credibility, and no credibility, P(C(x)). L )≥k) represents the water quality parameter x L It is a probability with a confidence level of not less than k. When k is 2, P(C(x)) L ≥2) indicates that the water quality parameter x L It is a probability with a confidence level of not less than 2, and its calculation formula is: P(C(x L )≥2)=P(C(x L )=3)+P(C(x L )=2) P(E=1,C(x L )≥k) represents the water quality parameter x L The reliability of the measured value is no less than k and the probability of water quality exceeding the standard occurring simultaneously; Among a set of measured values for water quality parameter x L in which the confidence level of the measured values is not less than k, the conditional probability of a water quality exceeding standard event occurring is expressed as P(E = 1丨C(x L ) ≥ k), according to the calculation formula of conditional probability P(E = 1 | C(x L ) ≥ k) = P(E = 1, C(x L ) ≥ k) / P(C(x L ) ≥ k) The algorithm for determining whether an alarm is triggered at river water quality measurement point L due to the exceeding of water quality parameter x can be determined by the following method. The basic idea is that when the measured value of a water quality parameter with a certain degree of credibility exceeds the limit for a period of time, it is considered that a water pollution event caused by the exceeding of water quality parameter x has occurred at water quality measurement point L. Let x L (t i Let ) represent the measured value of water quality parameter x at the current moment, and m c m is the number of times that the measured value of water quality parameter x is not lower than k, indicating a confidence level of k. Ec I represents the number of times water quality parameter x exceeds the limit when the reliability of the measured value is not less than k. L δ represents the number of times the water quality parameter x exceeds the limit consecutively when the reliability of the measured value is not less than k. δ is a constant input by the user from the central control cloud platform. L This is the conditional probability threshold for a water quality parameter x to exceed the standard when the reliability of the measured value is not less than k. It is also a constant input by the user from the central control cloud platform. a. Assign values ​​to i and m c m Ec =0; bi = i + 1; c. When C(x) L (t i When ))≥k holds, m c =m c +1, proceed to step d; When C(x) L (t i If ))≥k is not true, proceed to step e; d. When x L (t i When the limit is exceeded, m Ec =m Ec +1, proceed to step e; When x L (t i If the value does not exceed the limit, proceed to step e; e. When the number of iterations i is greater than I L Then proceed to step f; When the number of iterations i is not greater than I L Then proceed to step a; f. Calculate P(E = 1 | C(x L )) ≥ k) = (m Ec / i) / (m c / i); g. When P(E = 1丨C(x L ) ≥ k) ≥ δ L At this time, an alarm is given on the total control cloud platform, and step b is entered; When P(E = 1|C(x L ) ≥ k) < δ L then enter step a.

8. The river water quality monitoring system based on redundant sensors according to claim 5, characterized in that, The algorithm in step four is as follows: For water quality parameter x, the measured values ​​and their reliability at multiple adjacent water quality measurement points at different times are stored in a four-dimensional array G, whose elements are represented by [Lj][t]. i ][x Lj (t i )][C(x Lj (t i ))] indicates that Lj represents the j-th water quality measurement point deployed, and t i Indicates the time when the measurement was performed, x Lj (t i ) represents the water quality parameter x at the j-th water quality measurement point at time t. i The measured value at time, C(x) Lj (t i )) represents the reliability of the measured value of water quality parameter x at the j-th water quality measurement point; by comparing the measured values ​​x at different water quality measurement points Lj Lj (t i Whether the national standard limit is exceeded, and whether a large-scale water pollution incident has occurred in the river section.

9. The river water quality monitoring system based on redundant sensors according to claim 5, characterized in that, The data processing server sends the calculation results of steps two, three, and four to the central control cloud platform wirelessly, and deletes the data every quarter or half a year.

10. The river water quality monitoring system based on redundant sensors according to claim 5, characterized in that, The central control cloud platform includes a system display screen, a display module, a control module, an alarm module, and an equipment maintenance module. The display module, control module, alarm module, and equipment maintenance module are connected to the system display screen via wiring harnesses and are interconnected to form the central control cloud platform. The central control cloud platform can receive water quality parameters and results sent by the storage edge server wirelessly, and can display the water quality parameters and their reliability of the corresponding water quality measurement points in the form of charts or numbers on the system display screen.

11. The river water quality monitoring system based on redundant sensors according to claim 10, characterized in that, The display module includes a graphical display module, a single-point query display module, and a multi-point query display module. The single-point query display module and the multi-point query display module can convert the change information of water quality parameters into graphical curves and display them on the system display screen.

12. The river water quality monitoring system based on redundant sensors according to claim 11, characterized in that, The graph has lines of four colors: green, blue, yellow, and red. Green lines represent water quality parameters with high confidence, blue lines represent water quality parameters with medium confidence, yellow lines represent water quality parameters with low confidence, and red lines represent water quality parameters with no confidence.

13. The river water quality monitoring system based on redundant sensors according to claim 10, characterized in that, The control module can issue commands to retrieve water quality parameters stored in the data processing server and to set commands for transmitters or smart water quality sensors. When the water quality sensor is a non-smart sensor, it can only receive setting commands from the transmitter. In this case, the central control cloud platform can only set the transmitter and not the water quality sensor.

14. The river water quality monitoring system based on redundant sensors according to claim 10, characterized in that, The equipment maintenance module can display the data processing server, data transmission station, and water quality sensor and their working status on the system display screen of the central control cloud platform.

15. The river water quality monitoring system based on redundant sensors according to claim 14, characterized in that, The system display screen allows users to select a water quality parameter. Clicking the selection box for the corresponding confidence level will display all sensors whose measured values ​​fall within the selected confidence level. When the reliability of the water quality measurement value at the water quality measurement point is 0, it indicates that all three sensors have large errors and should be maintained. When the reliability of the water quality measurement value at the water quality measurement point is 1, it indicates that two of the three sensors of the same manufacturer and model have a large error and should be maintained. When the reliability of the water quality measurement value at the water quality measurement point is 2, it indicates that the third sensor among the three sensors has a large error, and maintenance personnel can perform maintenance when passing by. When the reliability of the water quality measurement value at the water quality measurement point is 3, it means that all three sensors at the water quality measurement point are in good working condition and no maintenance is required.

16. The river water quality monitoring system based on redundant sensors according to claim 1, characterized in that, The river water quality monitoring system based on redundant sensors can be used for water quality monitoring of water supply networks and drainage networks.

Images