A water body transparency measurement system, method, device, medium, and program product

Abstract

The embodiment of the application provides a kind of water transparency measurement system, method, equipment, medium and program product, it is related to water quality detection technical field, the system includes: measuring equipment and electronic equipment;Measuring equipment is obtained by gathering in the process of sinking from water surface into the water to be measured water data sequence including multiple groups of water data, each group of water data includes the light intensity data and water depth data collected at the same time;Electronic equipment inputs water data sequence into water transparency calculation model, obtains the water transparency that water transparency calculation model is based on water data sequence output.In the above system, water transparency is obtained by the light intensity data and water depth data of objective measurement, and is calculated by water transparency calculation model, and does not need to rely on the subjective visual observation of measurement personnel, to ensure the objectivity and consistency of measurement process, and obtain accurate water transparency measurement result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water quality detection, in particular to a water transparency measurement system, method, device, medium and program product. BACKGROUND

[0002] The currently used water transparency measurement method is the Sechi disc measurement method: the Sechi disc is a round iron disc painted with black and white colors, during measurement, a measuring personnel passes a rope through the center hole of the Sechi disc, and gradually sinks the Sechi disc into the water body through the rope until the black and white boundary line on the Sechi disc can no longer be seen. At this time, the length of the rope below the water surface is the water transparency of the water body.

[0003] However, the above measurement method relies on the visual judgment of the measuring personnel, and factors such as the visual differences of different measuring personnel, the observation angle of the measuring personnel, and the shaking of the Sechi disc caused by water flow impact will all affect the visual judgment of the measuring personnel, which leads to a lack of objectivity and consistency in the measurement results, and accurate measurement results cannot be obtained. SUMMARY

[0004] The purpose of the embodiments of the present application is to provide a water transparency measurement system, method, device, medium and program product to obtain accurate water transparency measurement results. The specific technical solutions are as follows:

[0005] In a first aspect, the embodiments of the present application provide a water transparency measurement system, the system comprising: a measuring device and an electronic device, the measuring device comprising a light sensor and a water depth sensor;

[0006] The measuring device collects a first water body data sequence during sinking into the water body to be measured from the water surface, wherein the first water body data sequence comprises a plurality of groups of water body data, and each group of water body data comprises light intensity data and water depth data collected at the same time;

[0007] The electronic device obtains the first water body data sequence, inputs the first water body data sequence into a pre-trained water transparency calculation model, obtains a first water body transparency output by the water transparency calculation model based on the first water body data sequence, and determines a target water transparency of the water body to be measured based on the first water body transparency.

[0008] The water transparency calculation model is trained based on the sample water transparency of a sample water body and a sample water body data sequence of the sample water body collected by the measuring device.

[0009] Optionally,

[0010] The measuring device collects a second water body data sequence during the process of being retrieved from the water body to the water surface, wherein the second water body data sequence comprises a plurality of groups of water body data, and each group of water body data comprises light intensity data and water depth data collected at the same time;

[0011] The electronic device obtains the second water body data sequence, inputs the second water body data sequence into a pre-trained water body transparency calculation model, and obtains a second water body transparency output by the water body transparency calculation model based on the second water body data sequence.

[0012] The electronic device is specifically configured to determine a target water body transparency of the water body to be measured based on the first water body transparency and the second water body transparency.

[0013] Optionally, the water body transparency calculation model is trained in the following manner:

[0014] The measuring device collects a sample water body data sequence during the process of being submerged into a sample water body from the water surface and / or being retrieved from the sample water body to the water surface, wherein the sample water body data sequence comprises a plurality of groups of water body data, and each group of water body data comprises light intensity data and water depth data collected at the same time;

[0015] A sample water body transparency of the sample water body is obtained.

[0016] The sample water body data sequence is input into a to-be-trained model, and a predicted water body transparency output by the to-be-trained model based on the change of light intensity with water depth reflected by the sample water body data sequence is obtained.

[0017] Based on the difference between the sample water body transparency and the predicted water body transparency, the model parameters of the to-be-trained model are adjusted, and the step of inputting the sample water body data sequence into the to-be-trained model is returned until the to-be-trained model converges, and the to-be-trained model at this time is taken as a water body transparency calculation model.

[0018] Optionally, the water body transparency calculation model is trained in the following manner:

[0019] The measuring device collects a sample water body data sequence during the process of being submerged into a sample water body from the water surface and / or being retrieved from the sample water body to the water surface, wherein the sample water body data sequence comprises a plurality of groups of water body data, and each group of water body data comprises light intensity data and water depth data collected at the same time;

[0020] A real water depth value corresponding to preset light intensity data collected by the measuring device is obtained as a sample water body transparency of the sample water body.

[0021] input the sample water body data sequence into the to-be-trained model, so that the to-be-trained model predicts a target corresponding relationship between water depth data and water depth values based on the sample water body data sequence, and determines a water depth value corresponding to target water depth data as a predicted water body transparency of the sample water body according to the target corresponding relationship, where the target water depth data is water depth data corresponding to the preset light intensity data in the sample water body data sequence;

[0022] adjust model parameters of the to-be-trained model based on a difference between the sample water body transparency and the predicted water body transparency, and return to the step of inputting the sample water body data sequence into the to-be-trained model until the to-be-trained model converges, and take the to-be-trained model at this time as a water body transparency calculation model.

[0023] Optionally,

[0024] The electronic device obtains a measurement position of the to-be-measured water body, records a corresponding relationship between the target water body transparency and the measurement position, and displays the target water body transparency at the measurement position on a map.

[0025] Optionally, the measurement device comprises a mainboard, a battery for supplying power to the mainboard, and a packaging shell for packaging the mainboard and the battery;

[0026] The mainboard is integrated with a light sensor, a water depth sensor, and a communication module, wherein a pressure-sensitive diaphragm of the water depth sensor is located on an outer side of the packaging shell, electronic elements of the water depth sensor are located on an inner side of the packaging shell, the light sensor is located on an inner side of a transparent upper surface of the packaging shell, and the electronic device obtains a water body data sequence through the communication module;

[0027] The packaging shell is provided with a lifting hole for mounting a lifting rope.

[0028] Optionally, the mainboard is further integrated with at least one of a temperature sensor, a pH sensor, a dissolved oxygen sensor, a conductivity sensor, and a chlorophyll sensor.

[0029] In a second aspect, an embodiment of the present application provides a water body transparency measurement method, which comprises:

[0030] obtaining a first water body data sequence, wherein the first water body data sequence is collected by a measurement device during sinking into a to-be-measured water body from a water surface, and the first water body data sequence comprises a plurality of groups of water body data, and each group of water body data comprises light intensity data and water depth data collected at the same time;

[0031] input the first water body data sequence into a pre-trained water body transparency calculation model to obtain a first water body transparency output by the water body transparency calculation model based on the first water body data sequence, wherein the water body transparency calculation model is trained based on sample water body data sequence of a sample water body and sample water body transparency of the sample water body;

[0032] determine the target water body transparency of the water body to be measured based on the first water body transparency.

[0033] Optionally, the method further comprises:

[0034] obtain a second water body data sequence, wherein the second water body data sequence is obtained by the measuring device during the process of being retrieved from the water body to be measured to the water surface, and the second water body data sequence comprises a plurality of groups of water body data, and each group of water body data comprises light intensity data and water depth data collected at the same time;

[0035] input the second water body data sequence into the pre-trained water body transparency calculation model to obtain a second water body transparency output by the water body transparency calculation model based on the second water body data sequence;

[0036] The method further comprises:

[0037] determine the target water body transparency of the water body to be measured based on the first water body transparency and the second water body transparency.

[0038] Optionally, the water body transparency calculation model is trained in the following manner:

[0039] obtain a sample water body data sequence obtained by the measuring device during the process of being submerged into a sample water body from the water surface and / or being retrieved from the sample water body to the water surface, wherein the sample water body data sequence comprises a plurality of groups of water body data, and each group of water body data comprises light intensity data and water depth data collected at the same time;

[0040] obtain sample water body transparency of the sample water body;

[0041] input the sample water body data sequence into a to-be-trained model to obtain a predicted water body transparency output by the to-be-trained model based on the change of light intensity with water depth reflected by the sample water body data sequence;

[0042] adjust model parameters of the to-be-trained model based on the difference between the sample water body transparency and the predicted water body transparency, and return to the step of inputting the sample water body data sequence into the to-be-trained model until the to-be-trained model converges, and take the to-be-trained model at this time as the water body transparency calculation model.

[0043] Optionally, the water transparency calculation model is trained in the following manner:

[0044] A sample water body data sequence collected by the measuring device during the process of sinking into the sample water body from the water surface and / or recovering to the water surface from the sample water body is obtained, wherein the sample water body data sequence includes a plurality of groups of water body data, and each group of water body data includes illumination intensity data and water depth data collected at the same time;

[0045] A real water depth value corresponding to preset illumination intensity data collected by the measuring device is obtained as a sample water transparency of the sample water body;

[0046] The sample water body data sequence is input into a to-be-trained model, so that the to-be-trained model predicts a target corresponding relationship between water depth data and water depth values based on the sample water body data sequence, and determines a water depth value corresponding to target water depth data as a predicted water transparency of the sample water body according to the target corresponding relationship, wherein the target water depth data is water depth data corresponding to the preset illumination intensity data in the sample water body data sequence;

[0047] Based on a difference between the sample water transparency and the predicted water transparency, model parameters of the to-be-trained model are adjusted, and the step of inputting the sample water body data sequence into the to-be-trained model is returned until the to-be-trained model converges, and the to-be-trained model at this time is taken as a water transparency calculation model.

[0048] Optionally, after the target water transparency of the to-be-measured water body is determined based on the first water transparency, the method further includes:

[0049] A measurement position of the to-be-measured water body is obtained;

[0050] A corresponding relationship between the target water transparency and the measurement position is recorded;

[0051] The target water transparency is displayed at the measurement position on a map.

[0052] In a third aspect, an embodiment of the present application provides a measuring device, which includes a mainboard, a battery for supplying power to the mainboard, and a packaging shell for packaging the mainboard and the battery; the mainboard is integrated with an illumination sensor, a water depth sensor, and a communication module;

[0053] A pressure sensing diaphragm of the water depth sensor is located on an outer side of the packaging shell, and electronic elements of the water depth sensor are located on an inner side of the packaging shell;

[0054] The light sensor is located inside a transparent upper surface of the package housing.

[0055] The communication module is configured to support data transmission between the measurement device and other devices.

[0056] In a fourth aspect, an electronic device is provided, which includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus.

[0057] The memory is configured to store a computer program.

[0058] The processor is configured to execute the program stored in the memory, and implement the method of any one of the second aspect.

[0059] In a fifth aspect, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the method of any one of the second aspect.

[0060] In a sixth aspect, a computer program product is provided, which is run on a computer to cause the computer to execute the method of any one of the second aspect.

[0061] The embodiments of the present application have the following beneficial effects:

[0062] The water transparency measurement system provided by the embodiments of the present application includes a measurement device and an electronic device, the measurement device includes a light sensor and a water depth sensor; the measurement device collects a first water body data sequence in the process of sinking into a water body to be measured from the water surface, wherein the first water body data sequence includes a plurality of groups of water body data, and each group of water body data includes light intensity data and water depth data collected at the same time; the electronic device acquires the first water body data sequence, inputs the first water body data sequence into a pre-trained water transparency calculation model, and acquires a first water transparency output by the water transparency calculation model based on the first water body data sequence; wherein the water transparency calculation model is trained based on sample water transparency of a sample water body and sample water body data sequence of the sample water body collected by the measurement device.

[0063] In the above system, the measurement of water transparency relies on light intensity data and water depth data objectively measured by the sensor, rather than subjective visual judgment of the measurement personnel, thereby ensuring the objectivity and consistency of the measurement process, and further, after the above data is input into the pre-trained water transparency calculation model, the water transparency calculation model can accurately calculate the water transparency based on the relationship between the pre-learned water body data sequence and the water transparency, so that an accurate water transparency measurement result can be obtained.

[0064] In addition, in a water body with a large water flow speed, the lifting rope for sinking the Sechi disc into the water body will be inclined, which results in that the length of the lifting rope below the water surface is greater than the actual water depth of the Sechi disc, and further results in that the water transparency measured by the Sechi disc is larger. The water depth data in the present application is measured based on the water depth sensor, and even if the water flow speed is large and the lifting rope for sinking the measuring device into the water body is inclined, it will not affect the measurement of the water depth data, and will not affect the measurement of the light intensity data. Therefore, the water transparency measurement system provided in the embodiments of the present application can still accurately measure the water transparency in a water body with a large water flow speed.

[0065] Of course, implementing any product or method of the present application does not necessarily require all the advantages described above. BRIEF DESCRIPTION OF DRAWINGS

[0066] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other embodiments can also be obtained by those skilled in the art based on these drawings.

[0067] Figure 1 A structural schematic diagram of the water transparency measurement system provided in the embodiments of the present application;

[0068] Figure 2 A model training method based on the embodiment shown in Figure 1 A flowchart of another model training method based on the embodiment shown in

[0069] Figure 3 A flowchart of another model training method based on the embodiment shown in Figure 1 A flowchart of another model training method based on the embodiment shown in

[0070] Figure 4 A flowchart of another model training method based on the embodiment shown in Figure 1 A flowchart of another model training method based on the embodiment shown in

[0071] Figure 5 A structural schematic diagram of the measuring device provided in the embodiments of the present application;

[0072] Figure 6 A flowchart of the water transparency measurement method provided in the embodiments of the present application;

[0073] Figure 7 A structural schematic diagram of the electronic device provided in the embodiments of the present application. DETAILED DESCRIPTION

[0074] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art based on the present application are within the scope of protection of the present application.

[0075] In order to obtain accurate water transparency measurement results, the present application provides a water transparency measurement system, method, electronic device, measurement device, computer readable medium and computer program product. First, the water transparency measurement system provided by the present application will be introduced.

[0076] As shown in Figure 1 A water transparency measurement system includes a measurement device 101 and an electronic device 102. The measurement device 101 includes an illumination sensor and a water depth sensor.

[0077] The measurement device 101 collects a first water body data sequence during sinking into the water body to be measured from the water surface.

[0078] The first water body data sequence includes a plurality of groups of water body data, and each group of water body data includes illumination intensity data and water depth data collected at the same time.

[0079] The electronic device 102 obtains the first water body data sequence, inputs the first water body data sequence into a pre-trained water transparency calculation model, obtains a first water transparency output by the water transparency calculation model based on the first water body data sequence, and determines a target water transparency of the water body to be measured based on the first water transparency.

[0080] The water transparency calculation model is trained based on sample water transparency of a sample water body and a sample water body data sequence collected by the measurement device from the sample water body.

[0081] The water transparency measurement system provided by the embodiments of the present application relies on the light intensity data and the water depth data measured by the sensor, instead of the subjective visual judgment of the measurement personnel, so as to ensure the objectivity and consistency of the measurement process. Furthermore, after the above data is input into the pre-trained water transparency calculation model, the water transparency calculation model can accurately calculate the water transparency based on the relationship between the pre-learned water data sequence and the water transparency, so that the accurate water transparency measurement result can be obtained. In addition, in a water body with a large flow velocity, the lifting rope for sinking the Secchi disc into the water body will be inclined, which results in that the length value of the lifting rope below the water surface is greater than the actual water depth of the Secchi disc, and further results in that the water transparency measured by the Secchi disc is too large. However, the water depth data in the present application is measured based on the water depth sensor, so even if the flow velocity is large and the lifting rope for sinking the measurement device into the water body is inclined, it will not affect the measurement of the water depth data, and will not affect the measurement of the light intensity data. Therefore, the water transparency measurement system provided by the embodiments of the present application can still accurately measure the water transparency in a water body with a large flow velocity.

[0082] The measurement device can include a light sensor and a water depth sensor. When it is necessary to measure the water transparency of the water body to be measured, the light sensor and the water depth sensor in the measurement device can be turned on, and then the measurement device can be slowly sunk into the water body to be measured by means of manual control or device control.

[0083] For example, the measurement personnel can manually release the lifting rope connected to the measurement device to slowly sink the measurement device into the water body to be measured. Alternatively, the lifting rope connected to the measurement device can be connected to a winch installed on a device that can float on the water surface, so that the lifting rope can be automatically released by the winch to slowly sink the measurement device into the water body to be measured.

[0084] During the sinking of the measurement device into the water body to be measured, the light sensor and the water depth sensor in the measurement device can continuously and synchronously collect data, so as to obtain the light intensity data and the water depth data before entering the water, and the light intensity data and the water depth data at different water depths of the water body to be measured after entering the water, that is, to obtain each group of water body data (each group of water body data includes the light intensity data and the water depth data collected at the same time), and then the data sequence composed of each group of water body data is the first water body data sequence of the water body to be measured.

[0085] After the measuring device completes data collection, the electronic device can establish a communication connection with the measuring device through a wired connection or a wireless connection. Then, the electronic device can obtain the first water body data sequence collected by the measuring device during the process of sinking into the water body to be measured, and then input the first water body data sequence into the pre-trained water body transparency calculation model.

[0086] It should be noted that if the measuring device stops sinking only after sinking to a shallow depth, the water body data collected by the measuring device will have the problem of insufficient data amount, and the water body transparency cannot be accurately determined based on the water body data sequence composed of each group of water body data. Therefore, the measuring device can be slowly retrieved from the water body to be measured to the water surface when the sinking depth of the measuring device meets the requirement of "accurately determining the water body transparency".

[0087] The requirement that the sinking depth of the measuring device meets the requirement of "accurately determining the water body transparency" can be achieved by, but not limited to, the following ways: the measuring personnel cannot observe the indicator light on the measuring device, or the sinking depth of the measuring device reaches a preset depth, or the length of the lifting rope is set to a preset length, and then the lifting rope is completely released when the measuring device is sunk into the water body to be measured.

[0088] The principle of calculating the water body transparency based on the water body data sequence by the water body transparency calculation model is briefly described as follows:

[0089] When light enters the water body, it will be absorbed and scattered by water molecules, plankton, suspended particles, etc. with the increase of depth. Therefore, the light intensity will show a decay trend with the increase of water depth, and the more turbid the water quality (i.e. the lower the water transparency), the faster the light intensity decays.

[0090] On this basis, the water body transparency calculation model can be pre-trained according to the "sample water body data sequence of the sample water body collected by the measuring device" and the "sample water body transparency of the sample water body", so that the water body transparency calculation model can learn the relationship between the "water body data sequence" and the "water body transparency".

[0091] Then, in the actual measurement process, after the electronic device inputs the first water body data sequence into the pre-trained water body transparency calculation model, the water body transparency calculation model can predict the water body transparency corresponding to the "first water body data sequence", i.e. the first water body transparency, according to the pre-learned relationship between the "water body data sequence" and the "water body transparency".

[0092] The electronic device can directly take the first water transparency as the target water transparency of the water body to be measured, or can obtain the target water transparency based on the first water transparency and other indicators, or can obtain a correction coefficient according to environmental information (light intensity on the water surface, wind level, etc.) when measuring, and then take the product of the correction coefficient and the first water transparency as the target water transparency, which is not limited here.

[0093] In the scheme provided by the embodiments of the present application, on the one hand, the measurement of the water transparency relies on the light intensity data and the water depth data measured by the sensor objectively, rather than the subjective visual judgment of the measurement personnel, thereby ensuring the objectivity and consistency of the measurement process. Further, after the above data is input into the pre-trained water transparency calculation model, the water transparency calculation model can accurately calculate the water transparency based on the relationship between the pre-learned water data sequence and the water transparency, so that an accurate water transparency measurement result can be obtained. On the other hand, in a water body with a large flow velocity, the lifting rope for sinking the Secchi disc into the water body will be inclined, which results in that the length value of the lifting rope below the water surface is greater than the actual water depth of the Secchi disc, and further results in that the water transparency measured by the Secchi disc is larger. However, the water depth data in the present application is measured based on the water depth sensor, so even if the lifting rope for sinking the measurement device into the water body is inclined due to a large flow velocity, it will not affect the measurement of the water depth data, and will not affect the measurement of the light intensity data. Therefore, the water transparency measurement system provided by the embodiments of the present application can still achieve accurate measurement of the water transparency in a water body with a large flow velocity.

[0094] As an embodiment of the present application, the water transparency measurement system can integrate the data collected by the measurement equipment in the sinking stage and the rising stage to calculate the water transparency. Specifically:

[0095] During the process of the measurement equipment being retrieved from the water body to be measured to the water surface, the light sensor and the water depth sensor in the measurement equipment can continuously and synchronously collect data, so as to obtain the light intensity data and the water depth data at different water depths of the water body to be measured, that is, to obtain each group of water body data (each group of water body data includes the light intensity data and the water depth data collected at the same time), and then the data sequence composed of each group of water body data is the second water body data sequence of the water body to be measured.

[0096] Similarly, after the measurement equipment collects data, the electronic device can also obtain the second water body data sequence through the communication connection between the electronic device and the measurement equipment, and input the second water body data sequence into the pre-trained water transparency calculation model, and then obtain the second water transparency output by the water transparency calculation model based on the second water body data sequence.

[0097] Because the first water body data and the second water body data are for the same water body and are collected at the same position of the water body, theoretically, the water body transparency reflected by the two is basically consistent, and on this basis, the electronic device can determine the target water body transparency of the water body to be measured based on the first water body transparency and the second water body transparency.

[0098] For example, the average of the two is taken as the target water body transparency, or the two are weighted and averaged because the data collected in the sinking process and the rising process have different reliabilities, and then the weighted average is taken as the target water body transparency, and the like.

[0099] In an implementation, before determining the target water body transparency based on the first water body transparency and the second water body transparency, the electronic device can first determine whether the difference between the first water body transparency and the second water body transparency is greater than a preset threshold value. If it is greater, it means that there is a large difference between the data collected by the measuring device in the rising stage and the sinking stage, and on this basis, the electronic device can output a prompt information to prompt that the data collection of the measuring device is wrong.

[0100] For example, the data in the sinking stage is normal, but the data in the rising stage is abnormally small because there is suspended matter above the measuring device at a certain depth range, so there is a large difference between the first water body transparency calculated according to the data in the sinking stage (i.e., the first water body data sequence) and the second water body transparency calculated according to the data in the rising stage (i.e., the second water body data sequence). At this time, the electronic device outputs a prompt information to prompt that the data collection of the measuring device is wrong.

[0101] It should be noted that in the process of sinking the measuring device from the water surface into the water body to be measured and retrieving the measuring device from the water body to be measured to the water surface, the measuring device may not distinguish between the water body data in the sinking stage and the water body data in the rising stage when storing the collected water body data, but stores the two as a whole water body data sequence. On this basis, after obtaining the whole water body data sequence, the electronic device can divide the water body data in the sinking stage and the water body data in the rising stage (i.e., divide the first water body data sequence and the second water body data sequence) from the whole water body data sequence through the change trend of the water depth data, that is, the part of the water body data sequence corresponding to the water depth data showing a continuous downward trend is the first water body data sequence, and the part of the water body data sequence corresponding to the water depth data showing a continuous upward trend is the second water body data sequence.

[0102] In the scheme provided by the embodiments of the present application, the water transparency measurement system can calculate the water transparency of the water body to be measured by comprehensively measuring the data collected by the measuring device in the sinking stage and the rising stage, so that the consistency of the water body data in the rising stage and the sinking stage is used to verify whether the data collected by the measuring device is reliable, thereby avoiding inaccurate water transparency calculated based on water body data with large errors.

[0103] The two ways of training the water transparency calculation model provided by the embodiments of the present application will be introduced below. Figure 2 and Figure 3 The two ways of training the water transparency calculation model provided by the embodiments of the present application will be introduced below.

[0104] As an implementation of the embodiments of the present application, the water transparency calculation model can be trained in the manner shown in Figure 2 .

[0105] S201, obtaining a sample water body data sequence collected by a measuring device in the process of sinking into a sample water body from the water surface and / or recovering to the water surface from the sample water body.

[0106] S202, obtaining a sample water transparency of the sample water body.

[0107] When training the model, a plurality of sample water bodies can be selected, and then for each sample water body, the sample water body data sequence of the sample water body is obtained by the measuring device in the embodiments of the present application, and the sample water transparency of the sample water body is obtained by a Sechi disk measurement method or the like. The way of obtaining the sample water body data sequence is similar to the way of obtaining the first water body data sequence and / or the second water body data sequence described above, and will not be repeated here.

[0108] It should be noted that when obtaining the sample water body data sequence of the sample water body, only the data collected by the measuring device in the process of sinking into the sample water body from the water surface can be obtained, only the data collected by the measuring device in the process of recovering to the water surface from the sample water body can be obtained, or the data collected by the measuring device in the process of sinking into the sample water body from the water surface and recovering to the water surface from the sample water body can be obtained.

[0109] S203, inputting the sample water body data sequence into the to-be-trained model to obtain a predicted water transparency output by the to-be-trained model based on the change of the illumination intensity with the water depth reflected by the sample water body data sequence.

[0110] After obtaining the sample water body data sequence, it can be input into the to-be-trained model, and then the to-be-trained model can predict the transparency of the sample water body based on the change of the illumination intensity with the water depth reflected by the sample water body data sequence.

[0111] S204, determining whether the to-be-trained model converges; if the determination result is no, performing step S205, and if the determination result is yes, performing step S206.

[0112] After obtaining the predicted water transparency output by the to-be-trained model, the electronic device can substitute the sample water transparency and the predicted water transparency into the pre-constructed loss function, and then determine whether the to-be-trained model converges based on the loss function. For example, after substituting the sample water transparency and the predicted water transparency into the pre-constructed loss function, if the obtained loss value is less than a preset loss value, it is determined that the to-be-trained model converges.

[0113] S205, adjusting the model parameters of the to-be-trained model based on the difference between the sample water transparency and the predicted water transparency.

[0114] If the determination result of step S204 is no, that is, the to-be-trained model does not converge, the model parameters of the to-be-trained model can be adjusted based on the difference between the sample water transparency and the predicted water transparency by using methods such as the stochastic gradient descent method, the momentum method, and the Adam (Adaptive Moment Estimation) method, and then returning to step S203 to repeat the training steps.

[0115] S206, taking the to-be-trained model at this time as the water transparency calculation model.

[0116] If the determination result of step S204 is yes, that is, the to-be-trained model converges, the to-be-trained model at this time can be taken as the water transparency calculation model, that is, the water transparency calculation model is trained.

[0117] In the scheme provided by the embodiments of the present application, the water transparency calculation model can be trained according to the water data sequence and the water transparency of each sample water body, so that the water transparency calculation model can learn the corresponding relationship between the change of the light intensity reflected by the water data sequence with the water depth and the water transparency. In this way, in the subsequent actual measurement process, the water transparency calculation model can automatically calculate the water transparency of the to-be-measured water body according to the pre-learned corresponding relationship after the water data sequence of the to-be-measured water body is input into the water transparency calculation model, so as to realize objective and accurate calculation of the water transparency and obtain accurate water transparency measurement results. Moreover, the water transparency measurement can be automatically adapted to different water bodies without manual adjustment of parameters.

[0118] As an implementation manner of the embodiments of the present application, the water transparency calculation model can be trained in the manner as shown in FIG. 8. Figure 3

[0119] ​S301, obtain a sample water body data sequence collected by the measuring device in the process of sinking into the sample water body from the water surface and / or recovering to the water surface from the sample water body.

[0120] S302, obtain a real water depth value corresponding to the target water depth data collected by the measuring device as the sample water body transparency of the sample water body.

[0121] When training the model, a plurality of sample water bodies can be selected, and then for each sample water body, the sample water body data sequence of the sample water body is obtained by the measuring device in the embodiment of the present application, and in the process of sinking the measuring device into the sample water body, the real water depth value corresponding to each time can be recorded by the laser depth finder and the like. Wherein, the way of obtaining the sample water body data sequence is similar to the way of obtaining the first water body data sequence and / or the second water body data sequence, which will not be repeated here.

[0122] It should be noted that when obtaining the sample water body data sequence of the sample water body, only the data collected by the measuring device in the process of sinking into the sample water body from the water surface can be obtained, only the data collected by the measuring device in the process of recovering to the water surface from the sample water body can be obtained, or the data collected by the measuring device in the process of sinking into the sample water body from the water surface and recovering to the water surface from the sample water body can be obtained.

[0123] At present, the measurement method of water body transparency is that the measuring personnel wear a hanging rope in the center hole of the Secchi disc, and gradually sink the Secchi disc into the water body from the water surface through the hanging rope, until the black and white boundary line on the Secchi disc can not be seen. At this time, the length value of the hanging rope below the water surface is the water body transparency of the water body. From the data quantization point of view, the visual critical state that "the measuring personnel can not see the black and white boundary line on the Secchi disc" corresponds to the light intensity of the position of the Secchi disc reducing to a certain lower threshold, which is insufficient to support the human eye to clearly distinguish the black and white boundary line on the Secchi disc. Therefore, this subjective visual critical state can be converted into a preset light intensity data, so as to realize the quantization of the measurement process.

[0124] On this basis, the real water depth value corresponding to the measuring device collecting the preset light intensity data can be determined based on the real water depth values recorded by the laser depth finder and the like, and the real water depth value is taken as the sample water body transparency of the sample water body.

[0125] S303, input the sample water body data sequence into the to-be-trained model, so that the to-be-trained model predicts the target corresponding relationship between the water depth data and the water depth value based on the sample water body data sequence, and determines the water depth value corresponding to the target water depth data as the predicted water body transparency of the sample water body according to the target corresponding relationship.

[0126] Although the water depth data in the sample water body data sequence can be directly converted to obtain the water depth value, different water body environments can affect the correspondence between the water depth data and the water depth value, which can result in a certain deviation between the water depth value obtained by directly converting the water depth data corresponding to the preset light intensity data and the real water depth value corresponding to the preset light intensity data measured by the measuring device.

[0127] For example, in the case of a water depth sensor being a pressure sensor, the water depth data is pressure data, but different water body environments can cause the pressure data and the water depth value not to strictly comply with the theoretical conversion of "1 bar ≈ 10.2 meters", so if the water depth data corresponding to the preset light intensity data is directly converted according to the above theoretical conversion, there will be a certain deviation between the converted water depth value and the real water depth value.

[0128] Therefore, in the embodiments of the present application, the model introduces light intensity data as a key basis for correcting the correspondence between the water depth data and the water depth value during training, specifically:

[0129] After inputting the sample water body data sequence into the to-be-trained model, the to-be-trained model calculates the light attenuation degree (current depth light intensity / surface light intensity), which directly reflects the light transmission characteristics of the water body (in turbid water, light will rapidly attenuate with increasing depth; in clear water, the attenuation is relatively slow).

[0130] On this basis, the to-be-trained model no longer relies on a fixed "water depth data-water depth value" correspondence, but dynamically predicts the real correspondence (i.e., the target correspondence) between "water depth data-water depth value" in the current sample water body environment by analyzing the light attenuation rate.

[0131] Finally, the to-be-trained model accurately converts the target water body data corresponding to the preset light intensity data in the sample water body data sequence to the water depth value according to the predicted target correspondence, and takes it as the predicted water body transparency of the sample water body.

[0132] S304, determine whether the to-be-trained model converges; if the determination result is no, execute step S305, and if the determination result is yes, execute step S306.

[0133] S305, adjust the model parameters of the to-be-trained model based on the difference between the sample water body transparency and the predicted water body transparency.

[0134] S306, take the to-be-trained model at this time as the water body transparency calculation model.

[0135] Steps S304-S306 are similar to steps S204-S206, and will not be described again here.

[0136] In the scheme provided by the embodiments of the present application, the water transparency calculation model can be trained according to the water body data sequence and the water transparency of each sample water body, so that the water transparency calculation model can accurately predict the corresponding relationship between the water body data in the current water body and the water depth value based on the light intensity data, so that in the subsequent actual measurement process, after the electronic device inputs the water body data sequence of the water body to be measured into the water transparency calculation model, the water transparency calculation model can call the “water depth data-water depth value” mapping relationship learned in the training and suitable for the current water body environment to determine the water transparency of the current water body, thereby realizing objective and accurate calculation of the water transparency, and obtaining accurate water transparency measurement results. And it can automatically adapt to the water transparency measurement of different water bodies without manual parameter adjustment.

[0137] In addition, in the process of training the model, in order to ensure the accuracy of the model obtained by training, and in the process of actual measurement based on the model, in order to ensure the accuracy of the water transparency output by the model, the following processing can be performed:

[0138] I. At least one of the following processes is performed on the water body data sequence.

[0139] 1. Missing value processing.

[0140] If the light intensity data / water depth data at a certain time in the water body data sequence is missing, the surrounding light intensity data / water depth data is used to fill it. For example, the mean of the light intensity data / water depth data of the five time points before and after this time point is used to fill the missing light intensity data / water depth data at this time point.

[0141] 2. Abnormal value filtering.

[0142] Based on physical knowledge: if the light intensity data is negative (sensor failure) and the water depth data exceeds the equipment range of the water depth sensor, the data is directly removed.

[0143] Based on statistical method: use “3σ principle” (data exceeding ±3 times of standard deviation of mean value is considered as abnormal) to filter the values with large fluctuations.

[0144] 3. Data standardization.

[0145] The light intensity data (for example, 0-10000 lux) and pressure value (for example, 0-30 bar) are normalized to the interval [0, 1] through the formula (x-min) / (max-min), so as to avoid the influence of the difference in the range of feature values on the model training (for example, the pressure value “bar” is much smaller than the light value “lux”, which causes the model to focus on the light feature).

[0146] 4. Rate of change of water depth.

[0147] Calculate the difference between the current water depth data and the water depth data at the previous moment to determine the instantaneous fluctuation of water depth. If the instantaneous fluctuation of water depth is large, combine it with the instantaneous fluctuation of light intensity data to determine whether the instantaneous fluctuation of water depth is a true change in water depth or an abnormal change in water depth caused by equipment vibration. If it is an abnormal change in water depth, then filter out the instantaneous fluctuation of water depth.

[0148] For example, if the water depth data is pressure data collected by a pressure sensor, then the difference (ΔP) between the current pressure and the pressure at the previous moment can be calculated to determine the instantaneous pressure fluctuation. If ΔP is too large, it may be caused by factors such as equipment vibration. It is necessary to combine the instantaneous fluctuation of the light intensity data to determine whether ΔP is the actual water depth change. If not, then ΔP should be filtered out.

[0149] II. Model Iterative Optimization.

[0150] Collect sample water data sequences and water transparency data, and periodically retrain the model with new sample data. If the accuracy is found to decrease in a certain scenario (such as highly turbid water), supplement the model with sample data for that scenario to iteratively train the model, improve the model's generalization ability, and avoid the problem of the model "working well in clear water but having large errors in turbid water".

[0151] As one embodiment of this application, after determining the transparency of the water body, the electronic device can display the water transparency at the corresponding location. Specifically:

[0152] When measuring water transparency, it is usually necessary to measure multiple points to avoid errors from a single measurement. In order to directly present the water transparency of the water body to be measured to the user, the electronic device can determine the measurement location of the target water transparency after acquiring the transparency of each target water body, that is, after obtaining the target water transparency of the water body at each location. Then, it can record the correspondence between the target water transparency and the measurement location for subsequent data retrieval. At the same time, it can also display the target water transparency at the measurement location on the map.

[0153] For example, targeting Figure 4 The lake shown is assumed to have water body data sequences collected at locations A, B, and C using measuring equipment. The electronic device calculates the target water transparency as 5m based on the data sequence at location A, 5.5m based on the data sequence at location B, and 3m based on the data sequence at location C. Then, the electronic device can... Figure 4The corresponding target water transparency is displayed at position A, position B and position C in the lake, respectively, to obtain Figure 4 The display effect shown.

[0154] In the scheme provided by the embodiments of the present application, after the electronic device calculates the target water transparency for the to-be-measured water body, the electronic device can display the target water transparency at the corresponding measurement position on the map. In this way, the user can intuitively view the water transparency at different positions of the to-be-measured water body through the display of the electronic device, so as to perform water monitoring, pollution source investigation, water treatment and the like.

[0155] The specific structure of the measurement device in the embodiments of the present application will be introduced as follows. Figure 5 As shown in the figure, the measurement device can include a mainboard 501, a battery 502 for supplying power to the mainboard, and a packaging shell 503 for packaging the mainboard and the battery. The mainboard is integrated with a light sensor, a water depth sensor and a communication module.

[0156] The pressure diaphragm of the water depth sensor is located on the outside of the packaging shell and directly contacts the to-be-measured water body, so as to measure the pressure value at the current water depth of the measurement device, and then convert the measured pressure value into water depth data based on the conversion relationship between pressure and water depth. The electronic elements of the water depth sensor are located on the inside of the packaging shell, so as to prevent the to-be-measured water body from damaging the electronic elements.

[0157] The light sensor is entirely located on the inside of the packaging shell, so as to prevent the to-be-measured water body from damaging the light sensor. Meanwhile, the light sensor is located on the inside of the transparent upper surface of the packaging shell, so as to enable the light sensor to measure the light intensity at the current water depth of the measurement device.

[0158] The communication module can support communication between the measurement device and the electronic device, so that the electronic device can obtain the water body data sequence collected by the measurement device.

[0159] The types and models of the above-mentioned packaging shell, water depth sensor, light intensity sensor and communication module and the like can be selected according to the actual use scene, and the embodiments of the present application do not make specific limitations thereon. For example, an acrylic plate can be selected as the packaging shell; an MS5837-30BA water depth sensor based on the pressure measurement principle, with a measurement range of 0-30 bar and a measurement accuracy of 0.2 cm can be selected; a BH1750 high-precision light sensor with high precision and low power consumption, which can accurately measure the change of light intensity at different depths of the water body can be selected; an HC-05 Bluetooth module can be selected as the communication module and the like.

[0160] In an embodiment, the package shell can further be provided with a lifting hole for mounting a lifting rope, so as to sink the measuring device into the water body to be measured based on the lifting rope, and to retrieve the measuring device from the water body to be measured to the water surface.

[0161] In an embodiment, the measuring device can further include a switch and a power indicator, wherein the measuring device can be turned on and off by the switch, and the power indicator can be turned on after the switch is turned on to indicate that the measuring device has been turned on. Further, the power indicator can be in a constant-on state when the measuring device is working, and in a flickering state when the measuring device is out of power or has a fault.

[0162] In an embodiment, the main board can further be integrated with at least one of a temperature sensor, a pH sensor, a dissolved oxygen sensor, a conductivity sensor, and a chlorophyll sensor in addition to the integrated light intensity sensor and water depth sensor, so that the measuring device can measure more water body parameters of the water body to be measured, so as to subsequently evaluate the quality of the water body to be measured based on the water body parameters measured by the measuring device.

[0163] In an embodiment, the battery of the measuring device can be a wireless charging battery, so that the measuring device can be replaced without disassembly, and on this basis, the measuring device can be in a completely closed state to avoid water leakage due to poor sealing.

[0164] In the scheme provided by the embodiments of the present application, the measuring device uses high-precision light intensity sensors and water depth sensors, which can accurately measure the slight changes in light intensity and water depth, and then based on the accurate light intensity data and water depth data, the electronic device can calculate more accurate water transparency. Compared with the traditional Sechi disc measurement method, the water transparency measurement accuracy is greatly improved. In combination with other water quality sensors, synchronous measurement and comprehensive analysis of multiple water body parameters can be realized, providing more data support for comprehensive evaluation of water quality. In addition, the high-strength pressure-resistant acrylic plate is used for packaging, so that the measuring device has good waterproof and pressure-resistant performance, and can work stably in water bodies of different depths and different water flow velocities.

[0165] The embodiments of the present application further provide a water transparency measurement method, as shown in Figure 6 The water transparency measurement method comprises the following steps.

[0166] S601, a first water body data sequence is acquired.

[0167] The first water body data sequence is acquired by the measuring device during sinking into the water body to be measured from the water surface, and the first water body data sequence includes multiple groups of water body data, and each group of water body data includes light intensity data and water depth data collected at the same time.

[0168] S602, input the first water body data sequence into the pre-trained water body transparency calculation model to obtain a first water body transparency output by the water body transparency calculation model based on the first water body data sequence.

[0169] The water body transparency calculation model is trained based on sample water body transparency of a sample water body and a sample water body data sequence of the sample water body collected by the measuring device.

[0170] S603, determine a target water body transparency of the water body to be measured based on the first water body transparency.

[0171] In the water body transparency measurement system provided by the embodiments of the present application, the measurement of water body transparency relies on the light intensity data and the water depth data objectively measured by the sensor, rather than the subjective visual judgment of the measurement personnel, thereby ensuring the objectivity and consistency of the measurement process. Furthermore, after the above data is input into the pre-trained water body transparency calculation model, the water body transparency calculation model can accurately calculate the water body transparency based on the relationship between the pre-learned water body data sequence and the water body transparency, and thus the accurate water body transparency measurement result can be obtained. In addition, in a water body with a large flow velocity, the lifting rope for sinking the Sechi disc into the water body will be inclined, which causes the length value of the lifting rope below the water surface to be greater than the actual water depth of the Sechi disc, and further causes the water body transparency measured by the Sechi disc to be larger. However, the water depth data in the present application is measured based on the water depth sensor, and even if the flow velocity is large and the lifting rope for sinking the measuring device into the water body is inclined, it will not affect the measurement of the water depth data, and will not affect the measurement of the light intensity data. Therefore, the water body transparency measurement system provided by the embodiments of the present application can still achieve accurate measurement of the water body transparency in a water body with a large flow velocity.

[0172] As an implementation manner of the embodiments of the present application, the above method can further include:

[0173] obtaining a second water body data sequence, wherein the second water body data sequence is collected by the measuring device in the process of being retrieved from the water body to be measured to the water surface, and the second water body data sequence includes a plurality of groups of water body data, and each group of water body data includes light intensity data and water depth data collected at the same time;

[0174] inputting the second water body data sequence into the pre-trained water body transparency calculation model to obtain a second water body transparency output by the water body transparency calculation model based on the second water body data sequence;

[0175] The above determining the target water body transparency of the water body to be measured based on the first water body transparency can include:

[0176] determine a target water transparency of the water body to be measured based on the first water transparency and the second water transparency.

[0177] As an implementation manner of the embodiment of the present application, the water transparency calculation model can be trained in the following manner:

[0178] obtain a sample water body data sequence collected by the measuring device in the process of sinking into the sample water body from the water surface and / or recovering to the water surface from the sample water body, wherein the sample water body data sequence comprises a plurality of groups of water body data, and each group of water body data comprises illumination intensity data and water depth data collected at the same time;

[0179] obtain a sample water transparency of the sample water body;

[0180] input the sample water body data sequence into the to-be-trained model, and obtain a predicted water transparency output by the to-be-trained model based on a change of the illumination intensity with the water depth reflected by the sample water body data sequence;

[0181] adjust model parameters of the to-be-trained model based on a difference between the sample water transparency and the predicted water transparency, and return to the step of inputting the sample water body data sequence into the to-be-trained model until the to-be-trained model converges, and take the to-be-trained model at this time as the water transparency calculation model.

[0182] As an implementation manner of the embodiment of the present application, the water transparency calculation model can be trained in the following manner:

[0183] obtain a sample water body data sequence collected by the measuring device in the process of sinking into the sample water body from the water surface and / or recovering to the water surface from the sample water body, wherein the sample water body data sequence comprises a plurality of groups of water body data, and each group of water body data comprises illumination intensity data and water depth data collected at the same time;

[0184] obtain a real water depth value corresponding to preset illumination intensity data collected by the measuring device as a sample water transparency of the sample water body;

[0185] input the sample water body data sequence into the to-be-trained model, so that the to-be-trained model predicts a target corresponding relationship between the water depth data and the water depth value based on the sample water body data sequence, and determines a water depth value corresponding to target water depth data as a predicted water transparency of the sample water body, wherein the target water depth data is water depth data corresponding to the preset illumination intensity data in the sample water body data sequence;

[0186] adjust model parameters of the to-be-trained model based on a difference between the sample water transparency and the predicted water transparency, and return to the step of inputting the sample water body data sequence into the to-be-trained model until the to-be-trained model converges, and take the to-be-trained model at this time as the water transparency calculation model.

[0187] As one embodiment of this application, after determining the target water body transparency based on the transparency of the first water body, the method may further include:

[0188] Obtain the measurement location of the water body to be tested;

[0189] Record the correspondence between the transparency of the target water body and the measurement location;

[0190] Display the transparency of the target water body at the measurement location on the map.

[0191] This application also provides an electronic device, such as... Figure 7 As shown, it includes a processor 701, a communication interface 702, a memory 703, and a communication bus 704, wherein the processor 701, the communication interface 702, and the memory 703 communicate with each other through the communication bus 704.

[0192] Memory 703 is used to store computer programs;

[0193] The processor 701, when executing the program stored in the memory 703, implements the water transparency measurement method described in any of the above embodiments.

[0194] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0195] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0196] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0197] The processor described above can be a general processor, including a central processing unit (CPU), a network processor (NP), etc.; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component.

[0198] In another embodiment provided in the present application, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the steps of the water transparency measurement method in any of the above embodiments.

[0199] In another embodiment provided in the present application, a computer program product is provided, and the computer program product includes instructions. When the computer program product is executed on a computer, the computer is caused to perform the water transparency measurement method in any of the above embodiments.

[0200] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available medium can be a magnetic medium (for example, floppy disk, hard disk, magnetic tape), an optical medium (for example, DVD), or a semiconductor medium (for example, solid state disk (SSD)) and the like.

[0201] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting; it is not intended to exclude myriad other embodiments of the present application that other present or future devices possess. That is, although specific embodiments have been disclosed herein, one of ordinary skill in the art will appreciate that other embodiments can be practiced under the teachings of the present application without departing from the true spirit and scope of the present application. Therefore, claims based on the disclosure as provided herein are contemplated as falling within the scope of the present application. Accordingly, the specification and figures are to be regarded in an illustrative manner and representations made herein should not be construed as being limiting to only the illustrative embodiments.

[0202] Each of the embodiments in the present specification is described in a related manner, and the same or similar parts among the embodiments can be referred to each other. Each of the embodiments focuses on the difference from other embodiments. In particular, for the method, electronic device, computer-readable storage medium, and computer program product embodiments, since they are basically similar to the system embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

[0203] The above only describes the preferred embodiments of the present application, and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A water transparency measurement system, characterized in that, The system includes: measuring equipment and electronic equipment, wherein the measuring equipment includes a light sensor and a water depth sensor; The measuring device collects a first water body data sequence during the process of sinking from the water surface into the water body to be measured. The first water body data sequence includes multiple sets of water body data, and each set of water body data includes light intensity data and water depth data collected at the same time. The electronic device acquires the first water body data sequence, inputs the first water body data sequence into a pre-trained water body transparency calculation model, acquires the first water body transparency output by the water body transparency calculation model based on the first water body data sequence, and determines the target water body transparency of the water body to be measured based on the first water body transparency. The water body transparency calculation model is trained based on the sample water body transparency of the sample water body and the sample water body data sequence of the sample water body collected by the measuring device. The water transparency calculation model was trained in the following manner: The measurement device acquires a sequence of sample water data during the process of sinking into the sample water body from the water surface and / or returning to the water surface from the sample water body. The sample water data sequence includes multiple sets of water data, each set of water data including light intensity data and water depth data acquired at the same time. The actual water depth value corresponding to the preset light intensity data collected by the measuring device is used as the water transparency of the sample water body. The sample water body data sequence is input into the model to be trained, so that the model to be trained calculates the light attenuation degree based on the sample water body data sequence, predicts the target correspondence between water depth data and water depth value according to the light attenuation degree, and determines the water depth value corresponding to the target water depth data according to the target correspondence, as the predicted water transparency of the sample water body, wherein the light attenuation degree is the ratio of the light intensity at the current depth to the light intensity at the water surface, and the target water depth data is the water depth data corresponding to the preset light intensity data in the sample water body data sequence; Based on the difference between the sample water transparency and the predicted water transparency, the model parameters of the model to be trained are adjusted, and the step of inputting the sample water data sequence into the model to be trained is returned until the model to be trained converges. The model to be trained at this time is then used as the water transparency calculation model.

2. The system according to claim 1, characterized in that, The measuring device collects a second water body data sequence during the process of returning from the water body to the water surface. The second water body data sequence includes multiple sets of water body data, and each set of water body data includes light intensity data and water depth data collected at the same time. The electronic device acquires the second water body data sequence, inputs the second water body data sequence into a pre-trained water body transparency calculation model, and obtains the second water body transparency output by the water body transparency calculation model based on the second water body data sequence. The electronic device is specifically used to determine the target water body transparency of the water body to be tested based on the transparency of the first water body and the transparency of the second water body.

3. The system according to claim 1, characterized in that, The electronic device acquires the measurement location of the water body to be measured, records the correspondence between the transparency of the target water body and the measurement location, and displays the transparency of the target water body at the measurement location on the map.

4. The system according to any one of claims 1-3, characterized in that, The measuring device includes: a motherboard, a battery for powering the motherboard, and a housing for encapsulating the motherboard and the battery; The motherboard integrates a light sensor, a water depth sensor, and a communication module. The pressure-sensitive diaphragm of the water depth sensor is located on the outside of the encapsulation shell, the electronic components of the water depth sensor are located on the inside of the encapsulation shell, and the light sensor is located on the inside of the transparent upper surface of the encapsulation shell. The electronic device acquires water body data sequences through the communication module. The outer casing is provided with lifting holes for installing lifting ropes.

5. The system according to claim 4, characterized in that, The motherboard also integrates at least one of the following: temperature sensor, pH sensor, dissolved oxygen sensor, conductivity sensor, and chlorophyll sensor.

6. A method for measuring water transparency, characterized in that, The method includes: Acquire a first water body data sequence, wherein the first water body data sequence is collected by the measuring device during the process of sinking from the water surface into the water body to be measured, and the first water body data sequence includes multiple sets of water body data, each set of water body data including light intensity data and water depth data collected at the same time; The first water body data sequence is input into a pre-trained water body transparency calculation model to obtain the first water body transparency output by the water body transparency calculation model based on the first water body data sequence. The water body transparency calculation model is trained based on the sample water body transparency of the sample water body and the sample water body data sequence of the sample water body collected by the measuring device. Based on the transparency of the first water body, the target water body transparency of the water body to be tested is determined; The water transparency calculation model was trained in the following manner: The measurement device acquires a sequence of sample water data during the process of sinking into the sample water body from the water surface and / or returning to the water surface from the sample water body. The sample water data sequence includes multiple sets of water data, each set of water data including light intensity data and water depth data acquired at the same time. The actual water depth value corresponding to the preset light intensity data collected by the measuring device is used as the water transparency of the sample water body. The sample water body data sequence is input into the model to be trained, so that the model to be trained calculates the light attenuation degree based on the sample water body data sequence, predicts the target correspondence between water depth data and water depth value according to the light attenuation degree, and determines the water depth value corresponding to the target water depth data according to the target correspondence, as the predicted water transparency of the sample water body, wherein the light attenuation degree is the ratio of the light intensity at the current depth to the light intensity at the water surface, and the target water depth data is the water depth data corresponding to the preset light intensity data in the sample water body data sequence; Based on the difference between the sample water transparency and the predicted water transparency, the model parameters of the model to be trained are adjusted, and the step of inputting the sample water data sequence into the model to be trained is returned until the model to be trained converges. The model to be trained at this time is then used as the water transparency calculation model.

7. The method according to claim 6, characterized in that, The method further includes: Acquire a second water body data sequence, wherein the second water body data sequence is collected by the measuring device during the process of returning from the water body to the water surface, and the second water body data sequence includes multiple sets of water body data, each set of water body data including light intensity data and water depth data collected at the same time; The second water body data sequence is input into a pre-trained water body transparency calculation model to obtain the second water body transparency output by the water body transparency calculation model based on the second water body data sequence. Determining the target water body transparency based on the transparency of the first water body includes: The target water body transparency is determined based on the transparency of the first water body and the transparency of the second water body.

8. The method according to claim 6, characterized in that, After determining the target water body transparency of the water body to be tested based on the transparency of the first water body, the method further includes: Obtain the measurement location of the water body to be tested; Record the correspondence between the transparency of the target water body and the measurement location; The transparency of the target water body is displayed at the measurement location on the map.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method of any one of claims 6-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 6-8.

11. A computer program product, characterized in that, When the computer program product is run on a computer, it causes the computer to perform the method described in any one of claims 6-8.

Images