A method, medium and system for analyzing the tendency error of seawater temperature, salinity and depth instrument

Abstract

The invention provides a seawater temperature-salinity-depth instrument tendency error analysis method, medium and system, belonging to the technical field of temperature-salinity-depth instruments, and comprising the following steps: collecting temperature, salinity and depth data measured by a plurality of temperature-salinity-depth instruments of the same batch in a sea area to be measured at different positions and at different time points; acquiring seawater flow data in the sea area to be measured, and establishing a seawater flow model; merging the data collected by each temperature-salinity-depth instrument into the seawater flow model, and using a Bayesian algorithm to find flying point data, wherein the flying point data is data that does not conform to statistical laws; arranging the flying point data of each temperature-salinity-depth instrument into a flying point matrix; constructing a flying point matrix with different single-cause errors based on historical data collected by the temperature-salinity-depth instrument with a single-cause error, which is recorded as an error flying point matrix set; matching the flying point matrix with the flying point matrix set, and using the single-cause error corresponding to the optimal matching result as the tendency error corresponding to the flying point matrix.

Description

Technical Field

[0001] The invention belongs to the technical field of temperature-salinity-depth instruments, and in particular, relates to a method, medium and system for analyzing the tendency error of a seawater temperature-salinity-depth instrument. Background Art

[0002] The temperature, salinity and depth instrument is one of the core equipment for ocean observation and is widely used in the fields of ocean survey, hydrological monitoring, climate research, etc. The long-term stable operation of the equipment and the output of accurate temperature, salinity and depth data are of great significance for the in-depth understanding and scientific management of the marine environment. At present, with the rapid development of my country's marine industry, various types of marine observation platforms and detection equipment are increasing. However, some equipment has large system errors due to failures or aging in long-term marine applications. This type of single error seriously affects the data quality and may cause users to make wrong scientific research conclusions or decision-making basis. In response to the stability problem of the temperature, salinity and depth instrument, the existing methods mainly rely on regular manual calibration and maintenance. However, the frequency of equipment recovery is low, and rapid error diagnosis at sea cannot be achieved. In addition, the number of equipment is large, the manual operation intensity is high, and the diagnostic efficiency is low. This has become a bottleneck for accurately obtaining marine observation data. In the prior art, the error judgment of the temperature, salinity and depth instrument is often based on manual experience or actual measurement and inspection of the temperature, salinity and depth instrument, which is inefficient. Summary of the invention

[0003] In view of this, the present invention provides a seawater temperature, salinity and depth meter tendency error analysis method, medium and system, which can solve the technical problem in the prior art that the error judgment of the temperature, salinity and depth meter is often based on manual experience or actual measurement and inspection of the temperature, salinity and depth meter, which is inefficient.

[0004] The present invention is achieved in that:

[0005] A first aspect of the present invention provides a method for analyzing the tendency error of a seawater temperature, salinity and depth instrument, which comprises the following steps:

[0006] S10, collecting temperature, salinity and depth data measured by multiple temperature, salinity and depth instruments of the same batch at different locations and time points in the sea area to be measured;

[0007] S20, obtaining seawater flow data of the sea area to be measured and establishing a seawater flow model;

[0008] S30, merging the data collected by each temperature-salinity-depth instrument into the seawater flow model, and using the Bayesian algorithm to find flying point data, wherein the flying point data is data that does not conform to statistical laws;

[0009] S40, arranging the flying point data of each temperature-salinity-depth instrument into a flying point matrix;

[0010] S50, based on the historical data collected by the temperature-salinity-depth instrument with a single-cause error, and using the method of steps S10 to S40, construct a flying point matrix with different single-cause errors, recorded as an error flying point matrix set;

[0011] S60, matching the flying point matrix with the flying point matrix set, and taking the single cause error corresponding to the best matching result as the tendency error corresponding to the flying point matrix.

[0012] On the basis of the above technical solution, the method for analyzing the tendency error of a seawater temperature, salinity and depth meter of the present invention can also be improved as follows:

[0013] Among them, the S10 step specifically includes: determining the scope of the sea area to be tested, deploying multiple temperature-salinity-depth meters at different locations in the sea area to be tested, deploying 2-3 temperature-salinity-depth meters at one depth, adjusting the sampling time interval of the temperature-salinity-depth meters to 1-10 minutes, and the duration covering the representative environmental change process; recovering the deep sampler, downloading the data, checking the data format and storage integrity, deleting obviously erroneous data, and performing data quality control.

[0014] Furthermore, the step S20 specifically includes: determining that the scale range of the model is consistent with the requirements of S10, collecting flow-related data including wind fields, waves, and tides to establish a hydrodynamic coupling model.

[0015] Furthermore, the S30 step specifically includes: setting mapping parameters, determining a mapping relationship considering model errors, adding position coordinates and time tags to the temperature, salinity and depth instrument data based on the mapping relationship, mapping the data to the model results, comparing the data with the model results, extracting statistical parameters, establishing a data anomaly detection model using Bayesian theory, and calculating the posterior probability to determine that the abnormal point is recorded as a flying point.

[0016] Further, the step S40 specifically includes: setting matrix rows and columns, the rows are temperature-salinity-depth meters, the columns represent abnormality types, determining the number of flying points in the data time span of more than one year, classifying them into corresponding error types and generating a thermal map matrix.

[0017] Furthermore, the step S50 specifically includes: acquiring historical data of single errors, processing to obtain a flying point matrix, optimizing a sample set for each single error matrix sample set, and considering the representativeness of different equipment types.

[0018] Furthermore, the step S60 specifically includes: importing the trained error matrix sample set, establishing a discrimination model to process the data to be tested to obtain a flying point matrix, inputting the matrix into the model, calculating the response recognition response of each error category, the maximum error type, and counting the discrimination accuracy of all samples, evaluating the model stability, and adding new samples to optimize the model.

[0019] The single cause error includes: sensor failure error, calibration error, and attachment error.

[0020] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the above-mentioned seawater temperature, salinity and depth meter tendency error analysis method.

[0021] A third aspect of the present invention provides a seawater temperature, salinity and depth meter tendency error analysis system, which includes the above-mentioned computer-readable storage medium.

[0022] Compared with the prior art, the beneficial effects of the method, medium and system for analyzing the tendency error of a seawater temperature, salinity and depth meter provided by the present invention are as follows: by using the ocean hydrodynamic numerical model to dynamically constrain the temperature and salinity data, various flying point situations can be effectively and automatically identified. Then, based on the flying point distribution matrix of the temperature, salinity and depth meter, a corresponding error knowledge base is established. For newly acquired data, its tendency error category is determined only by matrix matching. The rapid and automatic identification of single errors of the marine temperature, salinity and depth meter is achieved, with short delay, wide range and high accuracy; a large amount of ineffective manual maintenance work is avoided, and the efficiency is higher; the technical problem that the error judgment of the temperature, salinity and depth meter in the prior art is often based on manual experience or actual measurement and inspection of the temperature, salinity and depth meter is solved, which is inefficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0024] Figure 1 A flow chart of the method provided by the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0026] like Figure 1 As shown, it is a flow chart of a method for analyzing the tendency error of a seawater temperature, salinity and depth meter provided by the first aspect of the present invention, and the method comprises the following steps:

[0027] S10, collecting temperature, salinity and depth data measured by multiple temperature, salinity and depth instruments of the same batch at different locations and time points in the sea area to be measured;

[0028] S20, obtaining seawater flow data of the sea area to be measured and establishing a seawater flow model;

[0029] S30, merging the data collected by each temperature-salinity-depth instrument into the seawater flow model, and using the Bayesian algorithm to find the flying point data, which is the data that does not conform to the statistical law;

[0030] S40, arranging the flying point data of each temperature-salinity-depth instrument into a flying point matrix;

[0031] S50, based on the historical data collected by the temperature-salinity-depth instrument with a single-cause error, and using the method of steps S10 to S40, construct a flying point matrix with different single-cause errors, recorded as an error flying point matrix set;

[0032] S60, matching the flying point matrix with the flying point matrix set, and taking the single cause error corresponding to the optimal matching result as the tendency error corresponding to the flying point matrix.

[0033] The specific implementation methods of the above steps are described in detail below:

[0034] The specific implementation process of collecting the same batch of data from multiple temperature, salinity and depth instruments in the sea area to be measured in step S10 is as follows:

[0035] (1) Determine the scope of the sea area to be measured. This method is applicable to offshore and open sea environments, and the sea area range is 10-100 kilometers. It can be described using latitude and longitude coordinates or projection coordinates.

[0036] (2) The data of the temperature-salinity-depth instrument are those of the temperature-salinity-depth instruments deployed in the same batch;

[0037] (3) Deploy multiple temperature, salinity and depth meters in the sea area to be measured. The instrument location distance is representative and can collect environmental information at different locations. Generally, they are placed at three heights: surface, middle and bottom layers, with 2-3 instruments deployed at each depth.

[0038] (4) Adjust the sampling time interval and duration of the temperature, salinity and depth instrument. Considering the response time of water body changes, the interval is set to 1-10 minutes; the duration covers representative environmental change processes, such as tidal cycles.

[0039] (5) Retrieve the deep sampler and download real-time or offline data. Check the data format and storage to ensure integrity.

[0040] (6) Data quality control. Delete obviously erroneous data. Check time stamps and delete time jump data.

[0041] The specific implementation method of obtaining flow data and establishing a model in step S20 is as follows:

[0042] (1) Determine that the sea area scale for collecting model-related data is 10-100 kilometers, which is consistent with the sea area scale in step S10.

[0043] (2) Collect flow-related data, including wind fields, waves, tides, ocean currents, etc. The data source can be satellite remote sensing, automatic stations, and numerical model products.

[0044] (3) Establish a hydrodynamic coupling model for the marine environment. The VOF method, particle method, smooth particle method, etc. can be used to simulate flow fields and turbulence.

[0045] (4) Set the model boundary conditions, including wind stress, pressure field, etc., to drive the model movement.

[0046] (5) In addition to hydrodynamics, additional parameter models such as temperature and salinity can be coupled.

[0047] (6) Perform numerical calculations and output flow field results. Check quality, such as mass conservation, etc.

[0048] (7) Verify observation data. Extract site data and compare them with model results. Determine the model error range.

[0049] Step S30 is used to project the temperature-salinity-depth instrument data onto the flow model to realize anomaly recognition. Specifically, the data collected by each temperature-salinity-depth instrument is merged into the seawater flow model, and the Bayesian algorithm is used to find the flying point data, which is the data that does not conform to the statistical law, including the following specific steps:

[0050] (1) Set mapping parameters. Determine the time and space mapping relationship, taking into account the model error, generally taking 1.2 times the model error range.

[0051] (2) Data location tag: Add location coordinates and time tags to each temperature, salinity and depth instrument data.

[0052] (3) Based on the mapping relationship, the labeled data is mapped into the flow model field.

[0053] (4) Compare the data with the flow field results. Extract statistical parameters such as mean value, standard deviation, etc.

[0054] (5) Using Bayesian theory, a data anomaly probability test model is established. Assuming that the temperature-salinity-depth instrument data conform to the prior distribution of the flow field, the posterior probability is calculated.

[0055] (6) Set a threshold to determine whether the abnormal probability exceeds the threshold. Divide into abnormal flying points and normal points.

[0056] (7) Calculate the distribution of flying points and the proportion of total points at each measuring station to reflect the severity of the error.

[0057] The specific implementation method of arranging the flying point matrix in step S40 is as follows:

[0058] (1) Set the row and column components of the matrix. The rows of the matrix correspond to each temperature, salinity and depth instrument, and the columns represent various types of anomalies, such as sensor drift, residual thermal error, instrument reset, etc.

[0059] (2) Determine the data time span. It can be more than one year.

[0060] (3) Count the number of flying points of each temperature, salinity and depth instrument and classify them into the corresponding error type. Calculate the flying point ratio according to the thermal measurement counting method.

[0061] (4) Generate a thermal map matrix. The value of the matrix reflects the amount of error corresponding to the instrument, and the error characteristics can be judged intuitively.

[0062] (5) Repeat (1) to (4) to obtain heat maps at different time spans and observe the changing trend. You can also generate a dynamic map to display the results.

[0063] The method for simulating the error flying point matrix set based on historical data in step S50 is as follows:

[0064] (I) Obtain the historical observation data set of the temperature-salinity-depth instrument with a single error.

[0065] 1. Query the historical calibration data set from the measuring station and find the data of parameter drift.

[0066] 2. Retrieve the original data of the long-term deployed temperature, salinity and depth meters, and manually screen out the devices with single errors.

[0067] (ii) Process these historical data using steps S10 to S40 to obtain a flying point matrix corresponding to a single error. The matrix result is stored in the sample library.

[0068] (III) For various single errors, repeatedly obtain multiple sets of matrix samples and construct error matrix sets. The more matrix sets there are, the more accurate the error discrimination results will be.

[0069] (IV) Optimize the sample set content. Consider the representativeness and uniformity of samples based on different equipment types, years of service, initial accuracy, etc.

[0070] The implementation process of discriminating error categories based on matrix matching in step S60 is as follows:

[0071] (1) Establish an error discrimination model. Import the trained error matrix sample set, extract features for supervised learning, and obtain the discriminant function.

[0072] (2) Process the data of the temperature, salinity and depth meter to be measured into a flying point matrix format.

[0073] (3) Input the matrix into the discriminant model. The model calculates the projection response value of the matrix on each error category.

[0074] (4) The algorithm identifies the single error type with the largest response and outputs the identification result.

[0075] (5) Repeat the test multiple times and statistically analyze the results. Calculate the accuracy of all samples and evaluate the stability of the model.

[0076] (6) Connect to the database, feed back and store new samples and discrimination results, expand the sample library, and optimize the model online.

[0077] The following is a specific embodiment of the present invention:

[0078] The specific implementation process of collecting the same batch of data from multiple temperature, salinity and depth instruments in the sea area to be tested in step S10 is as follows:

[0079] (1) Determine the sea area range Ω to be measured and describe it using longitude and latitude coordinates:

[0080] Ω={(lat,lon)|lat min ≤lat≤lat max ,lon min ≤lon≤lon max}

[0081] Among them, lat and lon represent latitude and longitude coordinates respectively; lat min ,lat max ,lon min ,lon max Indicates the latitude and longitude range of the sea area to be measured. Ω indicates the range of the sea area to be measured, which is described as a two-dimensional area using longitude and latitude coordinates.

[0082] (2) Determine the set of temperature, salinity and depth instruments of the same batch S = {s 1 ,s 2 , ..., s N}, where N is the number of temperature, salinity and depth meters. i represents the i-th temperature-salinity-depth instrument.

[0083] (3) Select M positions P = {p 1 , p 2 , ..., p M} Place the temperature-salinity-depth instrument to meet the following requirements:

[0084] p i ∈Ω,i=1,2,...,M

[0085] Among them, M ≥ 3, position p i The coordinates in latitude and longitude are (lat i ,lon i ). P = {p 1 , p 2 , ..., pM} represents the set of temperature, salinity and depth instrument placement locations selected in the sea area Ω, where p i Represents the longitude and latitude coordinates of the i-th location.

[0086] (4) Set the sampling interval Δt and continuous sampling time T of the temperature, salinity and depth instrument:

[0087] Δt=1~10min,

[0088] T ≥ a representative environmental change cycle (such as tidal cycle)

[0089] (5) Recover the deep sampler and obtain the original observation data sequence of the temperature-salinity-depth instrument:

[0090] D={d 1 , d 2 , ..., d K}, where d k =(t k ,s m , T k , S k , H k ) represents the kth data, including the sampling time t k 、Temperature, salinity and depth instrument ID s m , Temperature T k , Salinity S k , depth H k ; where t k represents the sampling time, s m ∈S represents the corresponding temperature-salinity-depth instrument, T k , S k , H k Represents temperature, salinity and depth data respectively.

[0091] (6) Delete the obviously erroneous data from the original data D to obtain the quality-controlled data D′.

[0092] The specific implementation method of obtaining flow data and establishing a model in step S20 is as follows:

[0093] (1) Determine that the spatial range covered by the flow model is also Ω.

[0094] (2) Collect flow-related data, including:

[0095] W: wind field data; O: wave data; C: tide data; F: ocean current data.

[0096] (3) Establish a hydrodynamic model of the marine environment based on the NS equation and the continuity equation:

[0097]

[0098]

[0099] Among them, u represents the velocity vector field, ρ represents the seawater density, v represents the turbulent viscosity coefficient, p represents the pressure field, g represents the gravitational acceleration, and f represents the Coriolis force.

[0100] (4) Set the model boundary conditions: wind stress τ, pressure field distribution p, etc.

[0101] (5) The temperature T and salinity S transmission equations can be coupled to construct a temperature-salinity numerical model.

[0102] (6) Use numerical methods to solve the model equations, such as the finite difference method, and output the flow field results u′.

[0103] (7) Use the post-quality control observation data D′ to verify the model result u′ and determine the model error range ∈.

[0104] Specific implementation process of step S30:

[0105] (1) Set the mapping time error δt and spatial error δX, and consider the model error ∈:

[0106] δt=1.2∈ t ,δX=1.2∈ s

[0107] (2) Mark the time t′ of the observed data k and spatial coordinates (lat′ k , lon′ k ):

[0108] d′ k =(t′ k , lat′ k , lon′ k ,s m , T k , S k , H k )

[0109] (3) Based on the above error, d′ k Mapped to the model results:

[0110] d″ k =(t″ k ,x″ k ,s m , T k , S k , H k )

[0111] Among them, x″ k Represents the spatial coordinate in the flow model.

[0112] (4) At position x″ k Extract flow field data u″ k :

[0113] u″ k =u′(x″ k , t″ k )

[0114] (5) Establishing Bayesian test model:

[0115]

[0116] Among them, p(d″ k |u″ k ) represents data d″ k Relative to the flow field u″ k The probability of abnormality.

[0117] (6) Set the threshold θ, if p(d″) k |u″ k )<θ, then determine d″ k It is an abnormal flying point. Generally, θ=0.8 or determined based on experience.

[0118] (7) Count the percentage of flying points at each measuring station and evaluate the severity of the error.

[0119] The specific method of generating the flying point matrix in step S40 is as follows:

[0120] (1) Set the flying point matrix A∈R N×L , the row index corresponds to the temperature-salinity-depth instrument s i , the columns represent the exception types, and the number of columns is L.

[0121] (2) Set the data time span [t 0 , t 1 ].

[0122] (3) Statistical time span of each temperature, salinity and depth instrument i Number of flying points:

[0123] (d′ k ∈s i &&d′ k is the jth type of anomaly)

[0124] in, Represents the indicator function.

[0125] (4) Calculate the proportion of various abnormalities of each temperature, salinity and depth instrument:

[0126] Indicates the percentage of flying points corresponding to error type j of temperature, salinity and depth instrument i

[0127] (5) Obtain the heat map matrix:

[0128] A(i, j) = r ij

[0129] (6) Repeat the above steps to obtain matrices with different time spans.

[0130] In step S50, the error matrix set is obtained based on historical data:

[0131] (1) Collect historical data of temperature-salinity-depth meters with single errors.

[0132] (2) For each single error historical data, repeat S10-S40 to obtain the corresponding error matrix.

[0133] (3) Construct the error matrix sample set:

[0134] B={B 1 , B 2 , ..., B Q}, where Q is the number of samples.

[0135] (4) Optimize the representativeness of the sample set and increase the number of samples under different conditions.

[0136] Matrix matching discrimination error category in step S60:

[0137] (1) Import the matrix sample B and use the supervised learning method to train the discriminant model f().

[0138] (2) For new data, repeat S10-S40 to obtain the flying point matrix A′.

[0139] (3) Input into the discriminant model to obtain various error response values:

[0140]

[0141] (4) Output the category with the largest response as the recognition result, i.e., the tendency error:

[0142]

[0143] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for analyzing the tendency error of a seawater temperature, salinity and depth meter, characterized in that: The following steps are involved: S10, collecting temperature, salinity and depth data measured by multiple temperature, salinity and depth instruments of the same batch at different locations and time points in the sea area to be measured; S20, obtaining seawater flow data of the sea area to be measured and establishing a seawater flow model; S30, merging the data collected by each temperature-salinity-depth instrument into the seawater flow model, and using the Bayesian algorithm to find flying point data, wherein the flying point data is data that does not conform to statistical laws; S40, arranging the flying point data of each temperature-salinity-depth instrument into a flying point matrix; S50, based on the historical data collected by the temperature-salinity-depth instrument with a single-cause error, and using the method of steps S10 to S40, construct a flying point matrix with different single-cause errors, which is recorded as an error flying point matrix set; S60, matching the flying point matrix with the flying point matrix set, and taking the single cause error corresponding to the best matching result as the tendency error corresponding to the flying point matrix.

2. The method for analyzing the tendency error of a seawater temperature, salinity and depth meter according to claim 1, characterized in that: The S10 step specifically includes: determining the scope of the sea area to be tested, deploying multiple temperature-salinity-depth meters at different locations in the sea area to be tested, deploying 2-3 meters at one depth, adjusting the sampling time interval of the temperature-salinity-depth meters to 1-10 minutes, and the duration covers the representative environmental change process, recovering the deep sampler, downloading the data, checking the data format and storage integrity, deleting obviously erroneous data, and performing data quality control.

3. The method for analyzing the tendency error of a seawater temperature, salinity and depth meter according to claim 2 is characterized in that: The step S20 specifically includes: determining that the scale range of the model is consistent with the requirements of S10, collecting flow-related data including wind fields, waves, and tides, and establishing a hydrodynamic coupling model; wherein, determining that the scale range of the model is consistent with the requirements of S10 specifically refers to: determining that the scale of the sea area for collecting model-related data is 10-100 kilometers, which is consistent with the sea area range in step S10.

4. The method for analyzing the tendency error of a seawater temperature, salinity and depth meter according to claim 3 is characterized in that: The step S30 specifically includes: setting mapping parameters, determining mapping relationships by considering model errors, adding location coordinates and time tags to the temperature-salinity-depth instrument data, mapping the data to the model results based on the mapping relationships, comparing the data with the model results, extracting statistical parameters, establishing a data anomaly detection model using Bayesian theory, and calculating posterior probabilities to determine abnormal points, which are recorded as flying points.

5. The method for analyzing the tendency error of a seawater temperature, salinity and depth meter according to claim 4 is characterized in that: The step S40 specifically includes: setting matrix rows and columns, with rows representing temperature-salinity-depth instruments, columns representing abnormality types, determining a data time span of more than one year, counting the number of flying points, classifying them into corresponding error types, and generating a thermal map matrix; wherein, counting the number of flying points and classifying them into corresponding error types specifically refers to: counting the number of flying points of each temperature-salinity-depth instrument, classifying them into corresponding error types, and calculating the flying point ratio according to the thermal measurement counting method.

6. The method for analyzing the tendency error of a seawater temperature, salinity and depth meter according to claim 5, characterized in that: The step S50 specifically includes: acquiring historical data of single errors, processing to obtain a flying point matrix, optimizing a sample set for each single error matrix sample set, and considering the representativeness of different equipment types.

7. The method for analyzing the tendency error of a seawater temperature, salinity and depth meter according to claim 6 is characterized in that: The S60 step specifically includes: importing the trained error matrix sample set, establishing a discrimination model, processing the data to be tested to obtain a flying point matrix, inputting the matrix into the model, calculating the response in each error category, identifying the maximum error type of the response, counting the discrimination accuracy of all samples, evaluating the model stability, and adding samples to optimize the model.

8. The method for analyzing the tendency error of a seawater temperature, salinity and depth meter according to claim 1, characterized in that: The single cause errors include: sensor failure error, calibration error, and attachment error.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program instructions, which, when executed, are used to execute the method for analyzing the tendency error of a seawater temperature, salinity and depth meter according to any one of claims 1 to 8.

10. A seawater temperature, salinity and depth instrument tendency error analysis system, characterized in that: A computer-readable storage medium comprising the computer-readable storage medium of claim 9.

Images